Latest Codex app exposed a bug where redrawing the screen could cause speech to slip through even when it's turned off. Hopefully fixed now.

Run voice discovery, voice testing, clipboard I/O, and external scripts in isolated background worker lanes. Deliver results through Fenrir's event queue so manager state remains owned by the main loop.

AGENTS.md

@@ -0,0 +1,128 @@
+## Development Philosophy
+
+**Clean code over backward compatibility.** Breaking changes are acceptable in major versions.
+
+- PEP8 compliance (snake_case for all variables/functions/settings)
+- Remove deprecated code completely - no compatibility hacks
+- Trust users to update configs when upgrading
+
+**Exception**: Never break core accessibility (speech output, basic navigation).
+
+## Core Architecture
+
+### Drivers
+- **Screen**: `vcsaDriver` (TTY), `ptyDriver` (terminal emulation)
+- **Input**: `evdevDriver` (evdev), `x11Driver` (X11 terminal mode)
+- **Speech**: `speechdDriver` (speech-dispatcher), `genericDriver` (subprocess)
+- **Sound**: `gstreamerDriver` (GStreamer), `genericDriver` (Sox subprocess)
+- **Remote**: `unixDriver` (Unix sockets), `tcpDriver`
+
+### Command System
+Commands in trigger folders execute automatically:
+- `onKeyInput/`, `onCursorChange/`, `onScreenUpdate/`, `onScreenChanged/`, `onHeartBeat/`
+- Numeric prefixes control priority (lower = higher priority): `15000-char_echo.py`
+- Manual commands in `commands/commands/`
+
+### Core Managers (in `src/fenrirscreenreader/core/`)
+`fenrirManager.py`, `commandManager.py`, `eventManager.py`, `screenManager.py`, `inputManager.py`, `outputManager.py`
+
+### Configuration
+- Main: `config/settings/settings.conf`
+- Keyboards: `config/keyboard/` (desktop.conf, laptop.conf)
+- Sounds: `config/sound/`
+- Punctuation: `config/punctuation/`
+
+## 🚨 CRITICAL: Files to NEVER Modify
+
+**These are timing-critical - only fix critical bugs, never refactor:**
+- `evdevDriver.py` - Input event processing
+- `vcsaDriver.py` - Screen monitoring
+
+Symptoms of breakage: echo changes, input lag, missing screen updates, sync issues.
+
+## Development Commands
+
+```bash
+# Run from source (requires root)
+cd src/ && sudo ./fenrir -f -d   # foreground with debug
+
+# Validation before commit
+python3 tools/validate_syntax.py --fix
+python3 tools/validate_release.py --quick
+
+# Tests
+pytest tests/
+pytest tests/ --cov=src/fenrirscreenreader --cov-report=html
+```
+
+**Note**: Standard TTY Fenrir blocks - only one instance at a time. `fenrir -x`
+can run multiple foreground X terminal instances. Ask user to run tests that
+need real Fenrir instances or root access.
+
+## Creating Commands
+
+Use `commands/command_template.py` as base. Required methods:
+- `initialize(self, environment)`, `shutdown()`, `run()`, `get_description()`
+
+### Import Pattern (Required)
+Fenrir's command loader doesn't support relative imports:
+```python
+import os, importlib.util
+_base_path = os.path.join(os.path.dirname(__file__), 'base_file.py')
+_spec = importlib.util.spec_from_file_location("base_module", _base_path)
+_module = importlib.util.module_from_spec(_spec)
+_spec.loader.exec_module(_module)
+BaseClass = _module.BaseClass
+```
+
+### Consolidated Base Classes
+- **Bookmarks**: `bookmark_base.py` - `BookmarkCommand(id, 'read'|'set'|'clear')`
+- **Adjustments**: `adjustment_base.py` - `AdjustmentCommand('speech'|'sound', 'rate'|'pitch'|'volume', 'inc'|'dec')`
+- **VMenu Search**: `vmenu_search_base.py` - `VMenuSearchCommand('a'-'z')`
+
+## VMenu System
+
+Application-specific menus in `vmenu-profiles/KEY/{app}/`:
+- Directories = submenus, `.py` files = actions
+- Activate: `Fenrir+F10`, navigate with arrows, Enter to execute
+
+## Remote Control
+
+Unix socket: `/tmp/fenrirscreenreader-daemon.sock`
+TCP: localhost:22447
+
+```bash
+echo "command say Hello" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
+echo "ls" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
+echo "setting set speech#rate=0.8" | nc localhost 22447
+```
+
+In `fenrir -x`, each instance has a private socket:
+`/tmp/fenrirscreenreader-<pid>.sock`. One instance may also own the standard
+socket. Use `ls`/`list` on the standard socket to discover registered instances
+and their socket paths. Untargeted remote commands use a short cross-process
+claim lock so only one instance handles the same broadcast command; direct
+commands to a private instance socket still run on that instance.
+
+Disable in untrusted environments: `[remote] enable=False`
+
+## Key Implementation Details
+
+- Commands via `command_mod.command()` instantiation
+- Bookmark structure: `env['commandBuffer']['bookMarks'][ID][app]['1'/'2']`
+- VMenu search: `env['runtime']['vmenuManager'].searchEntry(char)`
+
+## Recent Fixes (Reference)
+
+- Enhanced application detection for screen/tmux sessions
+- Table navigation consistency (sound feedback)
+- Blank line detection in 22 command files
+- Character/space review during navigation
+
+## Safe vs Risky Changes
+
+**Safe**: Commands, VMenu profiles, config handling, sound/speech drivers, docs
+
+**Risky (extra testing)**: Core managers, event processing
+
+**Avoid**: evdevDriver, vcsaDriver, timing-critical code

README.md

@@ -8,7 +8,7 @@ This software is licensed under the LGPL v3.
 
 ## Key Features
 
-- **Multiple Interface Support**: Works in Linux TTY, and terminal emulators
+- **Linux Console Support**: Works as a Linux TTY screen reader, with optional X11 terminal mode
 - **Flexible Driver System**: Modular architecture with multiple drivers for speech, sound, input, and screen
 - **Review Mode**: Navigate and review screen content without moving the edit cursor
 - **Table Navigation**: Advanced table mode with column headers, cell-by-cell navigation, and boundary feedback
@@ -29,10 +29,8 @@ Fenrir is a Linux screen reader. Linux is the only officially supported platform
 
 **Other platforms (macOS, BSD, Windows):** Pull requests adding support for other operating systems may be accepted provided they do not break Linux functionality. However, no special care will be taken to preserve functionality on secondary platforms. If changes to Fenrir break support on a non-Linux OS, it is the responsibility of third-party contributors to submit fixes.
 
-- Linux (ptyDriver, vcsaDriver, evdevDriver) - Full support
-- macOS (ptyDriver) - Community-maintained, no guarantees
-- BSD (ptyDriver) - Community-maintained, no guarantees
-- Windows (ptyDriver) - Community-maintained, no guarantees
+- Linux TTY (`vcsaDriver`, `evdevDriver`) - Full support
+- X11 terminal emulators (`-x`, `ptyDriver`, `x11Driver`) - Supported
 
 
 ## Core Requirements
@@ -52,11 +50,8 @@ Fenrir is a Linux screen reader. Linux is the only officially supported platform
    - ReadWrite permission:
      - /dev/input
      - /dev/uinput
-2. **ptyDriver** - Terminal emulation input driver (cross-platform)
-   - python-pyte
-3. **atspiDriver** - AT-SPI input driver for desktop environments
-   - python-pyatspi2
-
+2. **x11Driver** - X11 terminal-scoped input driver for `fenrir -x`
+   - python-xlib
 ### Remote Drivers:
 1. **unixDriver** - Unix socket remote control (default)
    - socat (for command-line interaction)
@@ -73,7 +68,7 @@ Fenrir is a Linux screen reader. Linux is the only officially supported platform
      - /dev/tty[1-64]
      - /dev/vcsa[1-64]
      - read logind DBUS
-2. **ptyDriver** - Terminal emulation driver (cross-platform)
+2. **ptyDriver** - Terminal emulation screen driver for `fenrir -x`
    - python-pyte
   
 
@@ -84,7 +79,18 @@ Fenrir is a Linux screen reader. Linux is the only officially supported platform
    - python-speechd
 2. **genericDriver** - Generic subprocess speech driver
    - espeak or espeak-ng (or any TTS command)
-3. **debugDriver** - Debug speech driver for testing
+3. **dectalkDriver** - Serial DECtalk-compatible hardware speech driver
+   - RPITalk gadget mode or a DECtalk-compatible serial device
+4. **dectalkUsbDriver** - Native Access Solutions DECtalk USB driver
+   - USB ID `0dd0:2001`
+   - PyUSB and a libusb backend
+5. **litetalkDriver** - Serial LiteTalk-compatible hardware speech driver
+   - RPITalk gadget mode or a LiteTalk-compatible serial device
+6. **doubletalkDriver** - Serial DoubleTalk LT-compatible hardware speech driver
+   - DoubleTalk LT; does not support the internal DoubleTalk PC card
+7. **tripletalkDriver** - Serial TripleTalk-compatible hardware speech driver
+   - External DB9 serial TripleTalk devices, or USB models that expose a tty serial device
+8. **debugDriver** - Debug speech driver for testing
    - No dependencies
 
 
@@ -128,14 +134,24 @@ Requires root privileges
     sudo ./fenrir
 
 Settings are located in:
-- **After installation**: `/etc/fenrir/settings/settings.conf`
+- **System-wide installation**: `/etc/fenrirscreenreader/settings/settings.conf`
+- **User-local settings**: `~/.local/share/stormux/fenrirscreenreader/settings/settings.conf`
 - **Development**: `config/settings/settings.conf`
 
+When Fenrir runs as a regular user, including `fenrir -x`, it loads the
+user-local settings file if it exists. If it does not exist, Fenrir falls back
+to the system settings file, but saving settings writes to the user-local path.
+When Fenrir runs as root, settings save to the system-wide path by default.
+
+User sound themes are loaded before system themes from
+`~/.local/stormux/fenrir/sounds/`. User scripts are loaded before configured
+system scripts from `~/.local/stormux/fenrir/`.
+
 By default Fenrir uses:
 - **Sound driver**: genericDriver (via sox)
 - **Speech driver**: speechdDriver (via speech-dispatcher) 
-- **Input driver**: evdevDriver (Linux) or ptyDriver (other platforms)
-- **Screen driver**: vcsaDriver (Linux TTY) or ptyDriver (terminal emulation)
+- **Input driver**: evdevDriver for Linux TTY, x11Driver for `fenrir -x`
+- **Screen driver**: vcsaDriver for Linux TTY, ptyDriver for `fenrir -x`
 
 ## Audio Configuration
 
@@ -203,6 +219,7 @@ sudo /usr/share/fenrirscreenreader/tools/configure_pipewire.sh
 2. **Basic Navigation**:
    - **Fenrir Key**: By default `Insert`, `Keypad Insert`, or `Meta/Super` key
    - **Tutorial Mode**: `Fenrir + H` to learn all commands interactively
+   - **Speech History**: `Fenrir + Ctrl + H` to review recent speech
    - **Quit Fenrir**: `Fenrir + Q`
 
 3. **Essential Commands**:
@@ -222,7 +239,8 @@ Fenrir supports two main keyboard layouts:
 - **Desktop Layout**: Uses numeric keypad for navigation (recommended for desktop users)
 - **Laptop Layout**: Alternative bindings for keyboards without numeric keypad
 
-Configure in `/etc/fenrir/settings/settings.conf`:
+Configure in `/etc/fenrirscreenreader/settings/settings.conf` or your
+user-local settings file:
 ```ini
 [keyboard]
 keyboard_layout=desktop  # or 'laptop'
@@ -251,7 +269,8 @@ Fenrir includes a powerful remote control system that allows external applicatio
 
 ### Configuration
 
-Enable remote control in `/etc/fenrir/settings/settings.conf`:
+Enable remote control in `/etc/fenrirscreenreader/settings/settings.conf` or
+your user-local settings file:
 
 ```ini
 [remote]
@@ -266,7 +285,8 @@ enable_command_remote=True     # allow command execution
 ### Remote Drivers
 
 1. **unixDriver** (recommended): Uses Unix domain sockets
-   - Socket location: `/tmp/fenrirscreenreader-deamon.sock` (TTY mode) or `/tmp/fenrirscreenreader-<pid>.sock`
+   - Socket location: `/tmp/fenrirscreenreader-daemon.sock` for the standard control socket
+   - `fenrir -x` instances also create private sockets: `/tmp/fenrirscreenreader-<pid>.sock`
    - More secure, local-only access
    - Works with `socat`
 
@@ -279,89 +299,105 @@ enable_command_remote=True     # allow command execution
 
 The `socat` command provides the easiest way to send commands to Fenrir:
 
+#### Instance Discovery
+```bash
+# List registered Fenrir instances and their socket paths
+echo "ls" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
+```
+
+In X terminal mode (`fenrir -x`), multiple Fenrir instances can run at the same
+time. Each instance has a private socket, and one instance may also own the
+standard control socket. Use `ls` or `command ls` on the standard socket to find
+the private socket for a specific instance. Commands sent to the standard socket
+are handled by its owner when possible; otherwise they are forwarded to a
+registered private socket, preferring the sender's Fenrir ancestor when one can
+be found. Untargeted commands sent through a shared or broadcast path are
+claimed by one instance so duplicate instances do not all perform the same
+action.
+
 #### Basic Speech Control
 ```bash
 # Interrupt current speech
-echo "command interrupt" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
+echo "command interrupt" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
 
 # Speak custom text
-echo "command say Hello, this is a test message" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
+echo "command say Hello, this is a test message" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
 
 # Temporarily disable speech (until next keystroke)
-echo "command tempdisablespeech" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
+echo "command tempdisablespeech" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
 ```
 
 #### Settings Control
 ```bash
 # Enable highlight tracking mode
-echo "setting set focus#highlight=True" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
+echo "setting set focus#highlight=True" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
 
 # Change speech parameters
-echo "setting set speech#rate=0.8" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
-echo "setting set speech#pitch=0.6" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
-echo "setting set speech#volume=0.9" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
+echo "setting set speech#rate=0.8" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
+echo "setting set speech#pitch=0.6" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
+echo "setting set speech#volume=0.9" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
 
 # Change punctuation level (none/some/most/all)
-echo "setting set general#punctuation_level=all" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
-echo "setting set general#punctuation_level=none" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
+echo "setting set general#punctuation_level=all" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
+echo "setting set general#punctuation_level=none" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
 
 # Voice and TTS engine control
-echo "setting set speech#voice=en-us+f3" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
-echo "setting set speech#module=espeak-ng" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
+echo "setting set speech#voice=en-us+f3" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
+echo "setting set speech#module=espeak-ng" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
 
 # Disable sound temporarily
-echo "setting set sound#enabled=False" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
-echo "setting set sound#volume=0.5" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
+echo "setting set sound#enabled=False" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
+echo "setting set sound#volume=0.5" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
 
 # Keyboard and input settings
-echo "setting set keyboard#char_echo_mode=1" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
-echo "setting set keyboard#word_echo=True" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
+echo "setting set keyboard#char_echo_mode=1" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
+echo "setting set keyboard#word_echo=True" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
 
 # Screen control (ignore specific TTYs)
-echo "setting set screen#ignore_screen=1,2,3" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
+echo "setting set screen#ignore_screen=1,2,3" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
 
 # Multiple settings at once
-echo "setting set speech#rate=0.8;sound#volume=0.7;general#punctuation_level=most" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
+echo "setting set speech#rate=0.8;sound#volume=0.7;general#punctuation_level=most" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
 
 # Reset all settings to defaults
-echo "setting reset" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
+echo "setting reset" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
 
 # Save current settings
-echo "setting save" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
-echo "setting saveas /tmp/my-fenrir-settings.conf" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
+echo "setting save" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
+echo "setting saveas /tmp/my-fenrir-settings.conf" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
 ```
 
 #### Clipboard Operations
 ```bash
 # Place text into clipboard
-echo "command clipboard This text will be copied to clipboard" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
+echo "command clipboard This text will be copied to clipboard" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
 
 # Export clipboard to file
-echo "command exportclipboard" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
+echo "command exportclipboard" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
 ```
 
 #### Window Management
 ```bash
 # Define a window area (x1 y1 x2 y2)
-echo "command window 0 0 80 24" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
+echo "command window 0 0 80 24" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
 
 # Reset window to full screen
-echo "command resetwindow" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
+echo "command resetwindow" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
 ```
 
 #### VMenu Control
 ```bash
 # Set virtual menu context
-echo "command vmenu nano/file" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
+echo "command vmenu nano/file" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
 
 # Reset virtual menu
-echo "command resetvmenu" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
+echo "command resetvmenu" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
 ```
 
 #### Application Control
 ```bash
 # Quit Fenrir
-echo "command quitapplication" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
+echo "command quitapplication" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
 ```
 
 ### Using TCP Driver
@@ -406,6 +442,9 @@ setting <action> [parameters]
 **Application Commands:**
 - `command quitapplication` - Quit Fenrir
 
+**Instance Commands:**
+- `ls` / `list` / `command ls` / `command list` - List registered Fenrir instances and their socket paths
+
 #### Available Settings
 
 **Settings Commands:**
@@ -423,7 +462,22 @@ setting <action> [parameters]
 - `speech#volume=0.1-1.0` - Speech volume
 - `speech#voice=voice_name` - Voice selection (e.g., "en-us+f3")
 - `speech#module=module_name` - TTS module (e.g., "espeak-ng")
-- `speech#driver=driver_name` - Speech driver (speechdDriver/genericDriver)
+- `speech#driver=driver_name` - Speech driver (speechdDriver/genericDriver/dectalkDriver/dectalkUsbDriver/litetalkDriver/doubletalkDriver/tripletalkDriver)
+- `speech#hardware_device=/dev/ttyS0` - Hardware synth serial device for dectalkDriver/litetalkDriver
+- `speech#hardware_baud_rate=9600` - Hardware synth serial baud rate
+- `speech#history_size=50` - Number of spoken items kept in runtime speech history
+
+The native Access Solutions DECtalk USB, USB ID `0dd0:2001`, uses
+`dectalkUsbDriver`. It is autodetected and does not use
+`speech#hardware_device` or `speech#hardware_baud_rate`. Install PyUSB and a
+libusb backend first. From a source checkout, use
+`python3 -m pip install '.[dectalk_usb]'`; distributions may package PyUSB as
+`python3-usb` or `python-pyusb`. Root-run Fenrir can access the device directly.
+A non-root Fenrir process needs a local udev rule granting its account
+read/write access to `0dd0:2001`; do not make the device world-writable. Other
+USB hardware synths are supported only when Linux exposes them as a serial tty
+such as `/dev/ttyACM0` or `/dev/ttyUSB0`. Use an explicit
+`speech#hardware_device` path for serial hardware speech.
 - `speech#auto_read_incoming=True/False` - Auto-read new text
 
 *Sound Settings:*
@@ -552,7 +606,7 @@ Fenrir provides intelligent progress bar detection and audio feedback for variou
 
 To enable progress monitoring:
 1. Add a key binding in your keyboard layout file
-2. Or use the remote control system: `echo "command progress_bar_monitor" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock`
+2. Or use the remote control system: `echo "command progress_bar_monitor" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock`
 
 ### Progress Detection Patterns
 
@@ -609,11 +663,15 @@ Building...
 - **Non-blocking**: Progress tones don't interrupt speech or other functionality
 - **Configurable**: Can be enabled/disabled as needed
 
+Fenrir detects stable progress structures rather than application-specific
+status formats. Application-specific formats change too frequently to support
+reliably.
+
 ### Usage Examples
 
 ```bash
 # Enable progress monitoring
-echo "command progress_bar_monitor" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
+echo "command progress_bar_monitor" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
 
 # Common scenarios where progress monitoring is useful:
 wget https://example.com/large-file.zip    # Download progress
@@ -637,7 +695,7 @@ Progress monitoring can be configured through settings:
 #!/bin/bash
 # notify_fenrir.sh - Send notifications to Fenrir
 
-SOCKET="/tmp/fenrirscreenreader-deamon.sock"
+SOCKET="/tmp/fenrirscreenreader-daemon.sock"
 
 fenrir_say() {
     echo "command say $1" | socat - UNIX-CLIENT:$SOCKET
@@ -660,7 +718,7 @@ import os
 
 def send_fenrir_command(command):
     """Send command to Fenrir via Unix socket"""
-    socket_path = "/tmp/fenrirscreenreader-deamon.sock"
+    socket_path = "/tmp/fenrirscreenreader-daemon.sock"
     if os.path.exists(socket_path):
         sock = socket.socket(socket.AF_UNIX, socket.SOCK_STREAM)
         try:
@@ -692,8 +750,8 @@ send_fenrir_command("setting set speech#rate=0.9")
 
 **Commands not working:**
 - Verify `enable_command_remote=True` in settings
-- Check Fenrir debug logs: `/var/log/fenrir.log`
-- Test with simple command: `echo "command interrupt" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock`
+- Check Fenrir debug logs: `/tmp/fenrir.log`
+- Test with simple command: `echo "command interrupt" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock`
 
 ## Command Line Options
 
@@ -711,8 +769,8 @@ fenrir [OPTIONS]
 - `-o, --options SECTION#SETTING=VALUE;..` - Override settings file options
 - `-d, --debug` - Enable debug mode
 - `-p, --print` - Print debug messages to screen
-- `-e, --emulated-pty` - Use PTY emulation with escape sequences for input (enables desktop/X/Wayland usage)
-- `-E, --emulated-evdev` - Use PTY emulation with evdev for input (single instance)
+- `-x, --x11` - Use PTY emulation with X11 keyboard input scoped to the terminal window
+- `--x11-window-id WINDOWID` - X11 window id to use for `--x11` terminal mode
 - `-F, --force-all-screens` - Force Fenrir to respond on all screens, ignoring ignore_screen setting
 - `-i, -I, --ignore-screen SCREEN` - Ignore specific screen(s). Can be used multiple times. Combines with existing ignore settings.
 
@@ -721,8 +779,11 @@ fenrir [OPTIONS]
 # Run in foreground with debug output
 sudo fenrir -f -d
 
-# Use PTY emulation for desktop use
-sudo fenrir -e
+# Use PTY emulation with X11 terminal-scoped keybindings
+fenrir -x
+
+# Use X11 mode with an explicit terminal window id
+fenrir -x --x11-window-id 0x123456
 
 # Override settings via command line
 sudo fenrir -o "speech#rate=0.8;sound#volume=0.5"
@@ -735,6 +796,24 @@ sudo fenrir --ignore-screen 1
 sudo fenrir -i 1 -i 2    # Ignore screens 1 and 2
 ```
 
+### X11 Terminal Mode
+
+`fenrir -x` runs Fenrir inside a GUI terminal without root. It uses the PTY screen driver for terminal contents and the X11 input driver for Fenrir keybindings. Unlike evdev mode, it does not listen to the whole desktop; key grabs are scoped to the target terminal window so desktop screen readers can keep their own global bindings outside that terminal.
+
+By default, Fenrir targets the X11 window from `WINDOWID`, falling back to the active X11 window. If a terminal does not set `WINDOWID`, pass the window explicitly with `--x11-window-id`.
+
+X11 terminal mode uses the normal keyboard layout files, including desktop and laptop bindings. It supports Fenrir keys such as numpad Insert/`KEY_KP0`, CapsLock/`KEY_CAPSLOCK`, Insert/`KEY_INSERT`, and Super/Windows/`KEY_META`. Script keys such as Compose use `KEY_COMPOSE`.
+
+In this mode Fenrir runs as a regular user. Settings are loaded from
+`~/.local/share/stormux/fenrirscreenreader/settings/settings.conf` when present,
+otherwise from `/etc/fenrirscreenreader/settings/settings.conf`; saving settings
+creates or updates the user-local file. User sound themes and scripts are read
+first from `~/.local/stormux/fenrir/sounds/` and `~/.local/stormux/fenrir/`.
+
+This mode requires `python-xlib`.
+
+For users who want a dedicated PTY/terminal screen reader instead of Fenrir's Linux console focus, see TDSR: https://github.com/tspivey/tdsr
+
 ## Localization
 Translation files are located in the `locale/` directory. To install translations:
 

RELEASE_CHECKLIST.md

@@ -88,11 +88,11 @@ ls -la config/punctuation/default.conf
 # Test basic functionality (ask user to run)
 sudo ./src/fenrir --help
 
-# Test in emulation mode (safer for desktop environments)  
-sudo ./src/fenrir -e --version
+# Test version output
+./src/fenrir --version
 
-# Quick functionality test (3-5 seconds)
-sudo timeout 5 ./src/fenrir -e -f || echo "Timeout reached (expected)"
+# Quick console functionality test (3-5 seconds, ask user to run on a TTY)
+sudo timeout 5 ./src/fenrir -f || echo "Timeout reached (expected)"
 ```
 
 **Expected Result**: No immediate crashes, basic help/version output works

bugs

@@ -3,13 +3,13 @@ https://git.stormux.org/storm/fenrir/issues
 
 For bugs, please provide a debug file that shows the issue.
 How to create a debug file:
-1. first delete old stuff:
-sudo rm /var/log/fenrir.log
-2. start fenrir in debug mode
+1. start fenrir in debug mode
 sudo fenrir -d
 <do your stuff>
-3. 
+2. 
 stop fenrir (fenrirKey + q)
-the debug file is in /var/log/fenrir.log
+the debug file is in /tmp/fenrir.log
+if another Fenrir debug instance is already using it, check /tmp/fenrir2.log,
+/tmp/fenrir3.log, etc.
 
 please be as precise as possible to make it easy to solve the problem.

check-dependencies.py

@@ -50,7 +50,13 @@ def check_dependency(dep: Dependency) -> bool:
 dependencyList = [
     # Core dependencies
     Dependency('FenrirCore', 'core', 'core', 
-              pythonImports=['daemonize', 'enchant', 'pyperclip', 'setproctitle']),
+              pythonImports=[
+                  'daemonize',
+                  'enchant',
+                  'pyperclip',
+                  'setproctitle',
+                  'xdg'
+              ]),
     
     # Screen drivers
     Dependency('DummyScreen', 'screen', 'dummyDriver'),
@@ -58,16 +64,13 @@ dependencyList = [
               pythonImports=['dbus'],
               devicePaths=['/dev/vcsa']),
     Dependency('PTY', 'screen', 'ptyDriver',
-              pythonImports=['pyte', 'xdg']),
+              pythonImports=['pyte']),
               
     # Input drivers
     Dependency('DummyInput', 'input', 'dummyDriver'),
     Dependency('DebugInput', 'input', 'debugDriver'),
     Dependency('Evdev', 'input', 'evdevDriver',
               pythonImports=['evdev', 'evdev.InputDevice', 'evdev.UInput', 'pyudev']),
-    Dependency('PTYInput', 'input', 'ptyDriver',
-              pythonImports=['pyte']),
-              
     # Sound drivers
     Dependency('DummySound', 'sound', 'dummyDriver'),
     Dependency('DebugSound', 'sound', 'debugDriver'),
@@ -82,11 +85,7 @@ dependencyList = [
     Dependency('Speechd', 'speech', 'speechdDriver',
               pythonImports=['speechd']),
     Dependency('GenericSpeech', 'speech', 'genericDriver',
-              checkCommands=['espeak-ng']),
-              
-    # Additional dependencies
-    Dependency('Pexpect', 'core', 'pexpectDriver',
-              pythonImports=['pexpect'])
+              checkCommands=['espeak-ng'])
 ]
 
 defaultModules = {
@@ -94,8 +93,7 @@ defaultModules = {
     'VCSA',
     'Evdev',
     'GenericSpeech',
-    'GenericSound',
-    'Pexpect'
+    'GenericSound'
 }
 
 def check_all_dependencies():

config/keyboard/Readme.md

@@ -7,10 +7,6 @@ This directory contains keyboard layout files for Fenrir screen reader.
 
 - **desktop.conf** - Desktop layout using numeric keypad (recommended)
 - **laptop.conf** - Laptop layout for keyboards without numeric keypad
-- **nvda-desktop.conf** - NVDA-compatible desktop layout
-- **nvda-laptop.conf** - NVDA-compatible laptop layout
-- **pty.conf** - PTY emulation layout for terminal use
-- **pty2.conf** - Alternative PTY emulation layout
 
 ## Key Features
 
@@ -30,13 +26,21 @@ This directory contains keyboard layout files for Fenrir screen reader.
 - **Exit review**: `Fenrir + Keypad .`
 - **Screen reading**: `Fenrir + Keypad 5` (current screen)
 
+### Speech History
+- **Open speech history**: `Fenrir + Ctrl + H`
+- **Navigate history**: `Up` and `Down`
+- **Copy current item**: `Enter`
+- **Exit speech history**: `Escape`
+
 ## Configuration
 
-To change keyboard layout, edit `/etc/fenrir/settings/settings.conf`:
+To change keyboard layout, edit `/etc/fenrirscreenreader/settings/settings.conf`
+or your user-local settings file at
+`~/.local/share/stormux/fenrirscreenreader/settings/settings.conf`:
 
 ```ini
 [keyboard]
-keyboardLayout=desktop  # or laptop, nvda-desktop, nvda-laptop, pty, pty2
+keyboard_layout=desktop  # or laptop
 ```
 
 ## Available Key Constants
@@ -582,4 +586,3 @@ BTN_TRIGGER_HAPPY37
 BTN_TRIGGER_HAPPY38
 BTN_TRIGGER_HAPPY39
 BTN_TRIGGER_HAPPY40
-

config/keyboard/desktop.conf

@@ -1,4 +1,5 @@
 KEY_FENRIR,KEY_H=toggle_tutorial_mode
+KEY_FENRIR,KEY_CTRL,KEY_H=speech_history
 KEY_CTRL=shut_up
 KEY_FENRIR,KEY_KP9=review_bottom
 KEY_FENRIR,KEY_KP7=review_top
@@ -14,15 +15,12 @@ KEY_FENRIR,KEY_ALT,KEY_2=present_last_line
 KEY_KP5=review_curr_word
 KEY_KP4=review_prev_word
 KEY_KP6=review_next_word
-KEY_FENRIR,KEY_SHIFT,KEY_KP5=review_curr_word_phonetic
-KEY_FENRIR,KEY_SHIFT,KEY_KP4=review_prev_word_phonetic
-KEY_FENRIR,KEY_SHIFT,KEY_KP6=review_next_word_phonetic
+2,KEY_KP5=review_curr_word_spell
+3,KEY_KP5=review_curr_word_phonetic
 KEY_KP2=review_curr_char
 KEY_KP1=review_prev_char
 KEY_KP3=review_next_char
-KEY_FENRIR,KEY_SHIFT,KEY_KP2=review_curr_char_phonetic
-KEY_FENRIR,KEY_SHIFT,KEY_KP1=review_prev_char_phonetic
-KEY_FENRIR,KEY_SHIFT,KEY_KP3=review_next_char_phonetic
+2,KEY_KP2=review_curr_char_phonetic
 KEY_FENRIR,KEY_CTRL,KEY_KP8=review_up
 KEY_FENRIR,KEY_CTRL,KEY_KP2=review_down
 KEY_FENRIR,KEY_KPDOT=exit_review
@@ -83,6 +81,7 @@ KEY_FENRIR,KEY_CTRL,KEY_P=toggle_punctuation_level
 KEY_FENRIR,KEY_RIGHTBRACE=toggle_auto_spell_check
 KEY_FENRIR,KEY_BACKSLASH=toggle_output
 KEY_FENRIR,KEY_CTRL,KEY_E=toggle_emoticons
+KEY_FENRIR,KEY_CTRL,KEY_D=toggle_diff_mode
 KEY_FENRIR,KEY_CTRL,KEY_SHIFT,KEY_E=cycle_key_echo
 key_FENRIR,KEY_KPENTER=toggle_auto_read
 KEY_FENRIR,KEY_CTRL,KEY_T=toggle_auto_time
@@ -131,4 +130,3 @@ KEY_FENRIR,KEY_F8=export_clipboard_to_x
 KEY_FENRIR,KEY_CTRL,KEY_DOWN=read_all_by_line
 KEY_FENRIR,KEY_CTRL,KEY_PAGEDOWN=read_all_by_page
 KEY_FENRIR,KEY_SHIFT,KEY_V=announce_fenrir_version
-KEY_FENRIR,KEY_LEFTCTRL,KEY_F4=cycle_keyboard_layout

config/keyboard/laptop.conf

@@ -1,4 +1,5 @@
 KEY_FENRIR,KEY_H=toggle_tutorial_mode
+KEY_FENRIR,KEY_CTRL,KEY_H=speech_history
 KEY_CTRL=shut_up
 KEY_FENRIR,KEY_SHIFT,KEY_O=review_bottom
 KEY_FENRIR,KEY_SHIFT,KEY_U=review_top
@@ -14,15 +15,12 @@ KEY_FENRIR,KEY_ALT,KEY_2=present_last_line
 KEY_FENRIR,KEY_K=review_curr_word
 KEY_FENRIR,KEY_J=review_prev_word
 KEY_FENRIR,KEY_L=review_next_word
-KEY_FENRIR,KEY_ALT,KEY_K=review_curr_word_phonetic
-KEY_FENRIR,KEY_ALT,KEY_J=review_prev_word_phonetic
-KEY_FENRIR,KEY_ALT,KEY_L=review_next_word_phonetic
+2,KEY_FENRIR,KEY_K=review_curr_word_spell
+3,KEY_FENRIR,KEY_K=review_curr_word_phonetic
 KEY_FENRIR,KEY_COMMA=review_curr_char
 KEY_FENRIR,KEY_M=review_prev_char
 KEY_FENRIR,KEY_DOT=review_next_char
-KEY_FENRIR,KEY_ALT,KEY_COMMA=curr_char_phonetic
-KEY_FENRIR,KEY_ALT,KEY_M=prev_char_phonetic
-KEY_FENRIR,KEY_ALT,KEY_DOT=next_char_phonetic
+2,KEY_FENRIR,KEY_COMMA=review_curr_char_phonetic
 KEY_FENRIR,KEY_CTRL,KEY_I=review_up
 KEY_FENRIR,KEY_CTRL,KEY_COMMA=review_down
 KEY_FENRIR,KEY_SLASH=exit_review
@@ -81,6 +79,7 @@ KEY_FENRIR,KEY_SHIFT,KEY_CTRL,KEY_P=toggle_punctuation_level
 KEY_FENRIR,KEY_RIGHTBRACE=toggle_auto_spell_check
 KEY_FENRIR,KEY_CTRL,KEY_SHIFT,KEY_ENTER=toggle_output
 KEY_FENRIR,KEY_SHIFT,KEY_E=toggle_emoticons
+KEY_FENRIR,KEY_CTRL,KEY_D=toggle_diff_mode
 KEY_FENRIR,KEY_CTRL,KEY_SHIFT,KEY_E=cycle_key_echo
 KEY_FENRIR,KEY_CTRL,KEY_T=toggle_auto_time
 KEY_FENRIR,KEY_Y=toggle_highlight_tracking
@@ -131,4 +130,3 @@ KEY_FENRIR,KEY_F8=export_clipboard_to_x
 KEY_FENRIR,KEY_CTRL,KEY_DOWN=read_all_by_line
 KEY_FENRIR,KEY_CTRL,KEY_PAGEDOWN=read_all_by_page
 KEY_FENRIR,KEY_SHIFT,KEY_V=announce_fenrir_version
-KEY_FENRIR,KEY_LEFTCTRL,KEY_F4=cycle_keyboard_layout

config/keyboard/pty.conf

@@ -1,89 +0,0 @@
-# This file contains terminal escape sequences as shortcut
-# It is used for PTY screen / Input driver (Terminal emulation)
-# ^[ is used as escape
-
-# f1 - fenrir help
-^[OP=toggle_tutorial_mode
-# double tap control+end read attributes
-2,^[[1;5F=attribute_cursor
-#=toggle_has_attribute
-# escape - stop speech
-^[=shut_up
-# context menu key - stop speech
-^[[29~=shut_up
-# alt+shift+down - review to bottom
-^[[1;4B=review_bottom
-# alt+shift_up - review to top
-^[[1;4A=review_top
-# alt+down - review current line
-^[[1;3B=review_curr_line
-# alt+left - review previous line
-^[[1;3D=review_prev_line
-# alt+right - review next line
-^[[1;3C=review_next_line
-# alt+shift+left - beginning of line
-^[[1;4D=review_line_begin
-# alt+shift+right - end of line
-^[[1;4C=review_line_end
-# control+down - review current word
-^[[1;5B=review_curr_word
-# control+left - review previous word
-^[[1;5D=review_prev_word
-# control+right - review next word
-^[[1;5C=review_next_word
-# shift+down - review current character
-^[[1;2B=review_curr_char
-# shift+left - review previous character
-^[[1;2D=review_prev_char
-# shift+right - review next character
-^[[1;2C=review_next_char
-# control+shift+down - current character phonetic
-^[[1;6B=curr_char_phonetic
-# control+shift+left - previous character phonetic
-^[[1;6D=prev_char_phonetic
-# control+shift+right - next character phonetic
-^[[1;6C=next_char_phonetic
-# f2 - toggle sound
-^[OQ=toggle_sound
-# f3 - toggle highlight tracking
-^[OR=toggle_highlight_tracking
-alt+f12 - quit fenrir
-^[[24;3~=quit_fenrir
-# alt+f12 - time
-^[[24;3~=time
-# 2,alt+f12 - date
-2,^[[24;3~=date
-# alt+[ - previous clipboard
-^[[=prev_clipboard
-# alt+] - next clipboard
-^[]=next_clipboard
-# control+f6 - Read current clipboard
-^[[17;5~=curr_clipboard
-# f6 - copy to clipboard
-^[[17~=copy_marked_to_clipboard
-# shift+f6 - clear clipboard
-^[[17;2~=clear_clipboard
-# f7 - paste clipboard
-^[[18~=paste_clipboard
-# alt+f8 - export clipboard to X
-^[[19;3~=export_clipboard_to_x
-# control+f8 - import clipboard from X
-^[[19;5~=import_clipboard_from_x
-# alt+f9 - export clipboard to file
-^[[20;3~=export_clipboard_to_file
-# control+f9 - import clipboard from file
-^[[20;5~=import_clipboard_from_file
-# shift+f5 - remove marks
-^[[15;2~=remove_marks
-# f5 - set mark
-^[[15~=set_mark
-# f8 - Last utterance to clipboard
-^[[19~=copy_last_echo_to_clipboard
-# lat+\ Toggle auto announcement of indentation
-^[\=toggle_auto_indent
-# alt+end - temperarily disable speech
-^[[1;3F=temp_disable_speech
-# control+end - toggle auto read
-^[[1;5F=toggle_auto_read
-# F12 - cycle keyboard layout
-^[[24~=cycle_keyboard_layout

config/keyboard/pty2.conf

@@ -1,46 +0,0 @@
-# This file contains terminal escape sequences as shortcut
-# It is used for PTY screen / Input driver (Terminal emulation)
-# ^[ is used as escape
-
-^[h=toggle_tutorial_mode
-^[/=shut_up
-^[[D=shut_up
-^[O=review_bottom
-^[U=review_top
-#^[[1;3B=review_curr_line
-^[i=review_curr_line
-^[u=review_prev_line
-^[o=review_next_line
-^[J=review_line_begin
-^[L=review_line_end
-^[j=review_line_first_char
-^[L=review_line_last_char
-^[k=review_curr_word
-^[j=review_prev_word
-^[l=review_next_word
-^[,=review_curr_char
-^[m=review_prev_char
-^[.=review_next_char
-^[<=curr_char_phonetic
-^[M=prev_char_phonetic
-^[>=next_char_phonetic
-^[OR=toggle_sound
-^[OS=toggle_speech
-^[8=toggle_highlight_tracking
-^[Q=quit_fenrir
-^[t=time
-^[T=date
-^[[5~=prev_clipboard
-^[[6~=next_clipboard
-^[C=curr_clipboard
-^[c=copy_marked_to_clipboard
-^[v=paste_clipboard
-^[[15~=import_clipboard_from_file
-^[X=remove_marks
-^[x=set_mark
-^[\=toggle_auto_indent
-^[B=copy_last_echo_to_clipboard
-# alt+f8 - export clipboard to X
-# ^[[19;3~=export_clipboard_to_x
-# # control+f8 - import clipboard from X
-# ^[[19;5~=import_clipboard_from_x

config/settings/settings.conf

@@ -9,7 +9,8 @@ driver=gstreamerDriver
 #driver=genericDriver
 
 # Sound themes. These are the pack of sounds used for sound alerts.
-# Sound packs may be located at /usr/share/sounds
+# Sound packs may be located at ~/.local/stormux/fenrir/sounds,
+# /usr/share/sounds/fenrir, or /usr/share/sounds/fenrirscreenreader.
 theme=default
 
 # Sound volume controls how loud the sounds for your selected soundpack are.
@@ -18,14 +19,14 @@ volume=0.7
 
 # shell commands for generic sound driver
 # the folowing variable are substituted
-# fenrirVolume = the current volume setting
-# fenrirSoundFile = the soundfile for an soundicon
-# fenrirFrequence = the frequency to play
-# fenrirDuration = the duration of the frequency
+# fenrir_volume = the current volume setting
+# fenrir_sound_file = the soundfile for an soundicon
+# fenrir_frequency = the frequency to play
+# fenrir_duration = the duration of the frequency
 # the following command is used to play a soundfile
-generic_play_file_command=play -q -v fenrirVolume fenrirSoundFile
+generic_play_file_command=play -q -v fenrir_volume fenrir_sound_file
 #the following command is used to generate a frequency beep
-generic_frequency_command=play -q -v fenrirVolume -n -c1 synth fenrirDuration sine fenrirFrequence
+generic_frequency_command=play -q -v fenrir_volume -n -c1 synth fenrir_duration sine fenrir_frequency
 
 # Enable progress bar monitoring with ascending tones by default
 progress_monitoring=True
@@ -34,9 +35,16 @@ progress_monitoring=True
 # Turn speech on or off:
 enabled=True
 
-# Select speech driver, options are speechdDriver or genericDriver:
+# Select speech driver, options are speechdDriver, genericDriver,
+# dectalkDriver, dectalkUsbDriver, litetalkDriver, doubletalkDriver,
+# or tripletalkDriver:
 driver=speechdDriver
 #driver=genericDriver
+#driver=dectalkDriver
+#driver=dectalkUsbDriver
+#driver=litetalkDriver
+#driver=doubletalkDriver
+#driver=tripletalkDriver
 
 # The rate selects how fast Fenrir will speak. Options range from 0, slowest, to 1.0, fastest.
 rate=0.5
@@ -69,12 +77,30 @@ volume=1.0
 # Select the language you want Fenrir to use.
 #language=en
 
+# Hardware speech synthesizer serial device.
+# Used by dectalkDriver, litetalkDriver, doubletalkDriver, and tripletalkDriver.
+# dectalkUsbDriver autodetects the Access Solutions device and ignores this.
+# USB serial devices are supported if Linux exposes them as /dev/ttyACM*
+# or /dev/ttyUSB*. USB-only synths with no tty device need a separate driver.
+# Set an explicit device for hardware speech.
+# Examples:
+# hardware_device=/dev/ttyACM0  # RPITalk USB gadget mode
+# hardware_device=/dev/ttyUSB0  # USB serial adapter
+# hardware_device=/dev/ttyS0    # built-in serial port
+hardware_device=/dev/ttyS0
+
+# Serial baud rate for hardware speech synthesizers.
+hardware_baud_rate=9600
+
 # Read new text as it happens?
 auto_read_incoming=True
 
 # Speak individual numbers instead of whole string.
 read_numbers_as_digits = False
 
+# Number of spoken items kept in runtime speech history.
+history_size=50
+
 # Flood control: batch rapid updates instead of speaking each one
 # Number of updates within rapid_update_window to trigger batching
 rapid_update_threshold=5
@@ -91,17 +117,17 @@ max_batch_lines=100
 # Only enable flood control if this many new lines appear in the window
 flood_line_threshold=500
 
-# genericSpeechCommand is the command that is executed for talking
+# generic_speech_command is the command that is executed for talking
 # the following variables are replaced with values 
-# fenrirText = is the text that should be spoken
-# fenrirModule = may be the speech module like used in speech-dispatcher, not every TTY need this
-# fenrirLanguage = the language 
-# fenrirVoice = is the current voice that should be used. Set the voice variable above.
+# fenrir_text = is the text that should be spoken
+# fenrir_module = may be the speech module like used in speech-dispatcher, not every TTY need this
+# fenrir_language = the language
+# fenrir_voice = is the current voice that should be used. Set the voice variable above.
 # the current volume, pitch and rate is calculated like this
 # value = min + settingValue * (min - max )
-# fenrirVolume = is replaced with the current volume
-# fenrirPitch = is replaced with the current pitch
-# fenrirRate = is replaced with the current speed (speech rate)
+# fenrir_volume = is replaced with the current volume
+# fenrir_pitch = is replaced with the current pitch
+# fenrir_rate = is replaced with the current speed (speech rate)
 generic_speech_command=espeak-ng -a fenrir_volume -s fenrir_rate -p fenrir_pitch -v fenrir_voice -- "fenrir_text"
 
 # min and max values of the TTS system that is used in generic_speech_command
@@ -131,6 +157,8 @@ driver=evdevDriver
 device=ALL
 # gives Fenrir exclusive access to the keyboard and lets it control keystrokes.
 grab_devices=True
+# Optional X11 target window id for x11Driver. Empty uses WINDOWID or active window.
+x11_window_id=
 # Ignore shortcut bindings and pass all keys through without processing Fenrir commands.
 # When True, Fenrir will only monitor screen content without intercepting keyboard input.
 # the current shortcut layout located in /etc/fenrirscreenreader/keyboard
@@ -158,7 +186,8 @@ double_tap_timeout=0.2
 # The default is 0, no logging.
 debug_level=0
 # debugMode sets where the debug output should send to:
-# debugMode=File writes to debug_file (Default:/tmp/fenrir-PID.log)
+# debugMode=File writes to debug_file (Default:/tmp/fenrir.log)
+# If the default log is already in use, Fenrir uses /tmp/fenrir2.log, etc.
 # debugMode=Print just prints on the screen
 debug_mode=File
 debug_file=
@@ -172,11 +201,6 @@ respect_punctuation_pause=True
 replace_undefined_punctuation_with_space=True
 # Pause speech briefly at newline characters for better readability
 new_line_pause=True
-number_of_clipboards=50
-# used path for "export_clipboard_to_file"
-# $user is replaced by username
-#clipboardExportPath=/home/$user/fenrirClipboard
-clipboard_export_path=/tmp/fenrirClipboard
 # Convert text emoticons like :) to descriptive text (e.g., "smiling face")
 emoticons=True
 # Define the Fenrir modifier key(s) - used to trigger Fenrir commands
@@ -199,7 +223,8 @@ date_format=%%A, %%B %%d, %%Y
 auto_spell_check=True
 # Language for spell checking (format: language_COUNTRY, e.g., en_US, en_GB, es_ES)
 spell_check_language=en_US
-# path for your scripts "script_keys" functionality
+# path for your scripts "script_keys" functionality.
+# User-local scripts in ~/.local/stormux/fenrir are loaded first.
 script_path=/usr/share/fenrirscreenreader/scripts
 # Override default commands or add custom commands without modifying Fenrir installation
 # Leave empty to use default commands only
@@ -226,12 +251,37 @@ has_attributes=True
 # Shell to use for PTY emulation mode (empty = use system default shell)
 # Examples: /bin/bash, /bin/zsh, /usr/bin/fish
 shell=
+# Diff review presentation mode:
+# speech = spoken feedback only
+# sound = sound cues only (falls back to speech when cue is unavailable)
+# both = speech and sound cues
+diff_presentation=both
+# Diff review verbosity:
+# compact = concise role/location feedback
+# verbose = include diff line content during navigation
+diff_verbosity=compact
+
+[clipboard]
+# Number of clipboard history entries Fenrir keeps.
+number_of_clipboards=50
+# used path for "export_clipboard_to_file"
+# $user is replaced by username
+#clipboard_export_path=/home/$user/fenrirClipboard
+clipboard_export_path=/tmp/fenrirClipboard
+# Keep Fenrir's clipboard history and the X clipboard synchronized.
+# In fenrir -x, an empty sync_display uses the current DISPLAY.
+# In console/TTY mode, set sync_display explicitly, for example :0.
+sync_enabled=False
+sync_display=
+sync_interval=0.5
 
 [focus]
 # Follow and announce text cursor position changes
 cursor=True
 # Follow and announce highlighted/selected text changes (useful in menus)
 highlight=False
+# Suppress generic incoming announcements for full-screen TUIs that redraw often
+tui=False
 
 [remote]
 enable=True

config/sound/default/soundicons.conf

@@ -1,8 +1,6 @@
 # Screen Reader Turned On or Off
 ScreenReaderOn='ScreenReaderOn.wav'
 ScreenReaderOff='ScreenReaderOff.wav'
-# PTY bypass
-PTYBypass='PTYBypass.wav'
 # Cancel the current command
 Cancel='Cancel.wav'
 # Accept command

config/sound/template/soundicons.conf

@@ -1,8 +1,6 @@
 # Screen Reader Turned On or Off
 ScreenReaderOn=''
 ScreenReaderOff=''
-# PTY bypass
-PTYBypass=''
 # Cancel the current command
 Cancel=''
 # Accept command

docs/development.txt

@@ -22,7 +22,7 @@ src/fenrirscreenreader/
 │   ├── onKeyInput/       # Key input hooks
 │   └── help/             # Tutorial system
 ├── drivers/              # Driver implementations
-│   ├── inputDriver/     # Input drivers (evdev, pty, atspi)
+│   ├── inputDriver/     # Input drivers (evdev, x11)
 │   ├── screenDriver/    # Screen drivers (vcsa, pty)
 │   ├── speechDriver/    # Speech drivers (speechd, generic)
 │   └── soundDriver/     # Sound drivers (generic, gstreamer)
@@ -36,8 +36,7 @@ Fenrir uses a pluggable driver architecture:
 
 1. **Input Drivers**: Capture keyboard input
    - evdevDriver: Linux evdev (recommended)
-   - ptyDriver: Terminal emulation
-   - atspiDriver: AT-SPI for desktop
+   - x11Driver: X11 terminal-scoped input
 
 2. **Screen Drivers**: Read screen content
    - vcsaDriver: Linux VCSA devices
@@ -83,7 +82,6 @@ Fenrir supports various event hooks:
 - **onCursorChange**: Triggered when cursor moves
 - **onScreenUpdate**: Triggered on screen content changes
 - **onKeyInput**: Triggered on key presses
-- **onByteInput**: Triggered on byte-level input
 - **onScreenChanged**: Triggered when switching screens
 
 ## Development Setup
@@ -116,7 +114,7 @@ sudo ./fenrir -f -d -p
 
 # Debug output goes to:
 # - Console (with -p flag)
-# - /var/log/fenrir.log
+# - /tmp/fenrir.log
 ```
 
 ## Creating Commands
@@ -386,4 +384,4 @@ current_line = lines[self.env['screen']['newCursor']['y']]
 - **Wiki**: https://git.stormux.org/storm/fenrir/wiki
 - **Issues**: Use repository issue tracker
 - **Community**: IRC irc.stormux.org #stormux
-- **Email**: stormux+subscribe@groups.io
+- **Email**: stormux+subscribe@groups.io

docs/diff_review_mode.md

@@ -0,0 +1,48 @@
+# Diff Review Mode
+
+Diff review mode provides read-only navigation for unified and classic diff files with speech-first output.
+
+## Quick Workflow
+
+1. Copy the full absolute path of a diff file to the Fenrir clipboard.
+2. Press `Fenrir + Ctrl + D` to toggle diff mode on.
+3. Review the diff using the keys below.
+4. Press `Esc` to leave diff mode.
+
+If the clipboard does not contain a full absolute file path, Fenrir announces an error and does not enter diff mode.
+
+## Keys (Active Only In Diff Mode)
+
+- `h` - Next hunk
+- `Shift + h` - Previous hunk
+- `f` - Next file section
+- `Shift + f` - Previous file section
+- `a` - Next added line
+- `Shift + a` - Previous added line
+- `d` - Next removed line
+- `Shift + d` - Previous removed line
+- `Up` - Previous line
+- `Down` - Next line
+- `Left` - Previous character
+- `Right` - Next character
+- `Ctrl + Left` - Previous word
+- `Ctrl + Right` - Next word
+- `s` - Diff summary
+- `F1` - Speak key help
+- `Esc` - Exit diff mode
+
+## Speech Behavior
+
+- Added and removed content is spoken as `Added:` and `Removed:` lines.
+- Marker-only lines are normalized for speech:
+  - `+++` is spoken as `added`
+  - `---` is spoken as `removed`
+- Classic diff markers are spoken in plain language:
+  - `17c17` -> `line 17 changed`
+  - `17d16` -> `line 17 deleted`
+  - `16a17` -> `line 17 added`
+
+## Notes
+
+- Diff mode is read-only and does not modify the diff file.
+- Normal Fenrir key bindings are restored when diff mode exits.

docs/fenrir.1

@@ -13,13 +13,16 @@ fenrir \- A modern, modular console screen reader for Linux
 .IR SECTION#SETTING=VALUE;.. ]
 .RB [ \-d ]
 .RB [ \-p ]
-.RB [ \-e ]
-.RB [ \-E ]
+.RB [ \-x ]
+.RB [ \-\-x11-window-id
+.IR WINDOWID ]
 .RB [ \-F ]
+.RB [ \-i
+.IR SCREEN ]
 
 .SH DESCRIPTION
 Fenrir is a modern, modular, flexible and fast console screen reader written in Python 3. 
-It provides spoken feedback for Linux console applications and supports multiple interface types including TTY, terminal emulators, and desktop environments.
+It provides spoken feedback for Linux console applications and supports Linux TTYs plus X11 terminal mode.
 
 Fenrir features a modular driver architecture supporting multiple speech synthesizers, sound systems, input methods, and screen reading techniques. It includes advanced features like review mode, multiple clipboards, spell checking, bookmarks, and configurable key bindings.
 
@@ -38,7 +41,7 @@ Run Fenrir in the foreground instead of as a daemon. Useful for debugging and te
 
 .TP
 .BR \-s ", " \-\-setting " " \fISETTING-FILE\fR
-Path to a custom settings file. Default is /etc/fenrir/settings/settings.conf.
+Path to a custom settings file. Default is /etc/fenrirscreenreader/settings/settings.conf for root and ~/.local/share/stormux/fenrirscreenreader/settings/settings.conf for regular-user saves.
 
 .TP
 .BR \-o ", " \-\-options " " \fISECTION#SETTING=VALUE;..\fR
@@ -47,27 +50,29 @@ Multiple settings can be separated by semicolons.
 
 .TP
 .BR \-d ", " \-\-debug
-Enable debug mode. Debug information will be logged to /var/log/fenrir.log.
+Enable debug mode. Debug information will be logged to /tmp/fenrir.log by
+default. If another Fenrir debug instance is already using it, Fenrir uses
+/tmp/fenrir2.log, /tmp/fenrir3.log, etc.
 
 .TP
 .BR \-p ", " \-\-print
 Print debug messages to screen in addition to logging them.
 
-.TP
-.BR \-e ", " \-\-emulated-pty
-Use PTY emulation with escape sequences for input. This enables usage in desktop/X11/Wayland environments and terminal emulators.
-
-.TP
-.BR \-E ", " \-\-emulated-evdev
-Use PTY emulation with evdev for input (single instance mode).
-
 .TP
 .BR \-F ", " \-\-force-all-screens
-Force Fenrir to respond on all screens, ignoring the ignoreScreen setting. This temporarily overrides screen filtering for the current session.
+Force Fenrir to respond on all screens, ignoring the ignore_screen setting. This temporarily overrides screen filtering for the current session.
 
 .TP
 .BR \-i ", " \-I ", " \-\-ignore-screen " \fISCREEN\fR"
-Ignore specific screen(s). Can be used multiple times to ignore multiple screens. This is equivalent to setting ignoreScreen in the configuration file and will be combined with any existing ignore settings.
+Ignore specific screen(s). Can be used multiple times to ignore multiple screens. This is equivalent to setting ignore_screen in the configuration file and will be combined with any existing ignore settings.
+
+.TP
+.BR \-x ", " \-\-x11
+Run in X11 terminal mode using PTY screen monitoring and X11 keyboard input scoped to the terminal window.
+
+.TP
+.BR \-\-x11-window-id " " \fIWINDOWID\fR
+Use a specific X11 terminal window id for X11 terminal mode.
 
 .SH KEY CONCEPTS
 
@@ -186,12 +191,24 @@ Announce marked text
 .SS Settings File
 The main configuration file is located at:
 .IP \[bu] 2
-.B /etc/fenrir/settings/settings.conf
+.B /etc/fenrirscreenreader/settings/settings.conf
 (system-wide installation)
 .IP \[bu] 2  
+.B ~/.local/share/stormux/fenrirscreenreader/settings/settings.conf
+(user-local settings)
+.IP \[bu] 2
 .B config/settings/settings.conf
 (development/git version)
 
+When Fenrir runs as a regular user, including fenrir -x, it loads the
+user-local settings file if it exists. If it does not exist, Fenrir falls back
+to the system settings file, but saving settings writes to the user-local path.
+When Fenrir runs as root, settings save to the system-wide path by default.
+
+User sound themes are loaded before system themes from
+~/.local/stormux/fenrir/sounds/. User scripts are loaded before configured
+system scripts from ~/.local/stormux/fenrir/.
+
 .SS Key Sections
 .TP
 .B [speech]
@@ -223,6 +240,16 @@ speechdDriver - Speech-dispatcher (recommended)
 .IP \[bu] 4  
 genericDriver - Command-line TTS (espeak, etc.)
 .IP \[bu] 4
+dectalkDriver - DECtalk-compatible serial hardware speech
+.IP \[bu] 4
+dectalkUsbDriver - Native Access Solutions DECtalk USB hardware speech
+.IP \[bu] 4
+litetalkDriver - LiteTalk-compatible serial hardware speech
+.IP \[bu] 4
+doubletalkDriver - DoubleTalk LT-compatible serial hardware speech
+.IP \[bu] 4
+tripletalkDriver - TripleTalk-compatible serial hardware speech
+.IP \[bu] 4
 debugDriver - Debug/testing
 
 .TP
@@ -234,14 +261,12 @@ gstreamerDriver - GStreamer-based
 .IP \[bu] 4
 debugDriver - Debug/testing
 
-.TP  
+.TP
 .B Input Drivers:
 .IP \[bu] 4
 evdevDriver - Linux evdev (recommended for Linux)
 .IP \[bu] 4
-ptyDriver - Terminal emulation (cross-platform)
-.IP \[bu] 4
-atspiDriver - AT-SPI for desktop environments
+x11Driver - X11 terminal-scoped input for fenrir -x
 
 .TP
 .B Screen Drivers:
@@ -267,17 +292,13 @@ Start Fenrir as a daemon with default settings.
 .B fenrir -f -d
 Run Fenrir in foreground with debug output.
 
-.TP
-.B fenrir -e
-Run Fenrir with PTY emulation for desktop/terminal use.
-
 .TP
 .B fenrir -o "speech#rate=0.8;sound#volume=0.5"
 Override speech rate and sound volume settings.
 
 .TP
 .B fenrir -F
-Force Fenrir to work on all screens, ignoring ignoreScreen settings.
+Force Fenrir to work on all screens, ignoring ignore_screen settings.
 
 .TP
 .B fenrir -s /home/user/my-fenrir.conf
@@ -288,61 +309,77 @@ Use a custom configuration file.
 Fenrir includes a powerful remote control system for automation and integration with external applications.
 
 .SS Configuration
-Enable remote control in /etc/fenrir/settings/settings.conf:
+Enable remote control in /etc/fenrirscreenreader/settings/settings.conf or the user-local settings file:
 
 .EX
 [remote]
 enable=True
 driver=unixDriver
-enableSettingsRemote=True
-enableCommandRemote=True
+enable_settings_remote=True
+enable_command_remote=True
 .EE
 
 .SS Using socat with Unix Sockets
 
+.TP
+.B Instance Discovery:
+.EX
+# List registered Fenrir instances and their socket paths
+echo "ls" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
+.EE
+
+In X terminal mode (fenrir -x), multiple Fenrir instances can run at the
+same time. Each instance has a private socket at
+/tmp/fenrirscreenreader-<pid>.sock, and one instance may also own the
+standard control socket. Use ls or "command ls" on the standard socket to
+find the private socket for a specific instance. Commands sent to the standard
+socket are handled by its owner when possible; otherwise they are forwarded to a
+registered private socket, preferring the sender's Fenrir ancestor when one can
+be found.
+
 .TP
 .B Basic Speech Control:
 .EX
 # Interrupt current speech
-echo "command interrupt" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
+echo "command interrupt" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
 
 # Speak custom text
-echo "command say Hello, this is a test" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
+echo "command say Hello, this is a test" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
 
 # Temporarily disable speech
-echo "command tempdisablespeech" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
+echo "command tempdisablespeech" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
 .EE
 
 .TP
 .B Settings Control:
 .EX
 # Enable highlight tracking
-echo "setting set focus#highlight=True" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
+echo "setting set focus#highlight=True" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
 
 # Change speech rate
-echo "setting set speech#rate=0.8" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
+echo "setting set speech#rate=0.8" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
 
 # Change punctuation level (none/some/most/all)  
-echo "setting set general#punctuationLevel=all" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
+echo "setting set general#punctuation_level=all" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
 
 # Voice and TTS control
-echo "setting set speech#voice=en-us+f3" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
+echo "setting set speech#voice=en-us+f3" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
 
 # Multiple settings at once
-echo "setting set speech#rate=0.8;sound#volume=0.7;general#punctuationLevel=most" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
+echo "setting set speech#rate=0.8;sound#volume=0.7;general#punctuation_level=most" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
 
 # Reset all settings
-echo "setting reset" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
+echo "setting reset" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
 .EE
 
 .TP
 .B Clipboard Operations:
 .EX
 # Add text to clipboard
-echo "command clipboard Text to copy" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
+echo "command clipboard Text to copy" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
 
 # Export clipboard to file
-echo "command exportclipboard" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
+echo "command exportclipboard" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
 .EE
 
 .SS Command Reference
@@ -355,6 +392,8 @@ command say <text> - Speak the specified text
 command interrupt - Stop current speech
 .IP \[bu] 2
 command tempdisablespeech - Disable speech until next key press
+.IP \[bu] 2
+ls / list / command ls / command list - List registered Fenrir instances
 
 .TP
 .B Settings Commands:
@@ -385,7 +424,7 @@ command quitapplication - Quit Fenrir
 .TP
 .B Speech Control:
 .IP \[bu] 2
-general#punctuationLevel=none/some/most/all - Punctuation verbosity
+general#punctuation_level=none/some/most/all - Punctuation verbosity
 .IP \[bu] 2
 speech#rate=0.1-1.0 - Speech rate
 .IP \[bu] 2
@@ -400,28 +439,32 @@ speech#module=module_name - TTS module
 .TP
 .B Screen Control:
 .IP \[bu] 2
-screen#ignoreScreen=1,2,3 - TTY screens to ignore
+screen#ignore_screen=1,2,3 - TTY screens to ignore
 .IP \[bu] 2
 focus#highlight=True/False - Follow text highlighting
 
 .TP
 .B Input Control:
 .IP \[bu] 2
-keyboard#charEchoMode=0-2 - Character echo mode
+keyboard#char_echo_mode=0-2 - Character echo mode
 .IP \[bu] 2
-keyboard#wordEcho=True/False - Word echo
+keyboard#word_echo=True/False - Word echo
 
 .SH FILES
 .TP
-.B /etc/fenrir/settings/settings.conf
+.B /etc/fenrirscreenreader/settings/settings.conf
 Main configuration file
 
 .TP
-.B /etc/fenrir/keyboard/
+.B ~/.local/share/stormux/fenrirscreenreader/settings/settings.conf
+User-local configuration file
+
+.TP
+.B /etc/fenrirscreenreader/keyboard/
 Keyboard layout definitions
 
 .TP
-.B /etc/fenrir/punctuation/
+.B /etc/fenrirscreenreader/punctuation/
 Punctuation level definitions
 
 .TP
@@ -429,11 +472,15 @@ Punctuation level definitions
 System sound themes
 
 .TP
-.B ~/.local/share/fenrirscreenreader/sounds/  
+.B ~/.local/stormux/fenrir/sounds/
 User sound themes
 
 .TP
-.B /var/log/fenrir.log
+.B ~/.local/stormux/fenrir/
+User scripts
+
+.TP
+.B /tmp/fenrir.log
 Debug log file
 
 .TP
@@ -516,4 +563,4 @@ This software is licensed under the LGPL v3.
 Full documentation: https://git.stormux.org/storm/fenrir/wiki
 
 .PP
-Support: stormux+subscribe@groups.io
+Support: stormux+subscribe@groups.io

docs/fenrir.adoc

@@ -27,9 +27,12 @@ and voices provided by Speech Dispatcher. +
 Keyboard shortcuts are bound to commands to read the screen, review it
 and perform other actions like modify the settings on the fly, switch
 languages, copy and paste text, provide miscellaneous information. +
-=== Configuration file === The permanent settings are recorded in the
-configuration file /etc/fenrirscreenreader/settings.conf, that is
-heavily commented. +
+=== Configuration file === The system-wide settings are recorded in the
+configuration file /etc/fenrirscreenreader/settings/settings.conf, that is
+heavily commented. Regular-user sessions, including `+fenrir -x+`, load
+`+~/.local/share/stormux/fenrirscreenreader/settings/settings.conf+` when it
+exists, otherwise they fall back to the system settings file. Saving settings
+as a regular user creates or updates the user-local file. +
 These settings are described in this user guide.
 
 To start Fenrir once type as root:
@@ -1214,17 +1217,11 @@ Enable debug mode. Debug information will be logged.
 `+-p, --print+`::
 Print debug messages to screen in addition to logging them.
 
-`+-e, --emulated-pty+`::
-Use PTY emulation with escape sequences for input. This enables usage in desktop/X11/Wayland environments and terminal emulators.
-
-`+-E, --emulated-evdev+`::
-Use PTY emulation with evdev for input (single instance mode).
-
 `+-F, --force-all-screens+`::
-Force Fenrir to respond on all screens, ignoring the ignoreScreen setting. This temporarily overrides screen filtering for the current session.
+Force Fenrir to respond on all screens, ignoring the ignore_screen setting. This temporarily overrides screen filtering for the current session.
 
 `+-i, -I, --ignore-screen <SCREEN>+`::
-Ignore specific screen(s). Can be used multiple times to ignore multiple screens. This is equivalent to setting ignoreScreen in the configuration file and will be combined with any existing ignore settings.
+Ignore specific screen(s). Can be used multiple times to ignore multiple screens. This is equivalent to setting ignore_screen in the configuration file and will be combined with any existing ignore settings.
 
 ==== Set settings options
 
@@ -1277,56 +1274,72 @@ enable_command_remote=True
 
 The `+socat+` command provides the easiest way to send commands to Fenrir:
 
+===== Instance Discovery
+
+....
+# List registered Fenrir instances and their socket paths
+echo "ls" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
+....
+
+In X terminal mode (`+fenrir -x+`), multiple Fenrir instances can run at the
+same time. Each instance has a private socket at
+`+/tmp/fenrirscreenreader-<pid>.sock+`, and one instance may also own the
+standard control socket. Use `+ls+` or `+command ls+` on the standard socket to
+find the private socket for a specific instance. Commands sent to the standard
+socket are handled by its owner when possible; otherwise they are forwarded to a
+registered private socket, preferring the sender's Fenrir ancestor when one can
+be found.
+
 ===== Basic Speech Control
 
 ....
 # Interrupt current speech
-echo "command interrupt" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
+echo "command interrupt" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
 
 # Speak custom text
-echo "command say Hello, this is a test message" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
+echo "command say Hello, this is a test message" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
 
 # Temporarily disable speech (until next keystroke)
-echo "command tempdisablespeech" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
+echo "command tempdisablespeech" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
 ....
 
 ===== Settings Control
 
 ....
 # Enable highlight tracking mode
-echo "setting set focus#highlight=True" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
+echo "setting set focus#highlight=True" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
 
 # Change speech rate
-echo "setting set speech#rate=0.8" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
+echo "setting set speech#rate=0.8" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
 
 # Change punctuation level (none/some/most/all)
-echo "setting set general#punctuation_level=all" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
+echo "setting set general#punctuation_level=all" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
 
 # Voice and TTS control
-echo "setting set speech#voice=en-us+f3" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
+echo "setting set speech#voice=en-us+f3" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
 
 # Multiple settings at once
-echo "setting set speech#rate=0.8;sound#volume=0.7;general#punctuation_level=most" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
+echo "setting set speech#rate=0.8;sound#volume=0.7;general#punctuation_level=most" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
 
 # Reset all settings to defaults
-echo "setting reset" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
+echo "setting reset" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
 ....
 
 ===== Clipboard Operations
 
 ....
 # Place text into clipboard
-echo "command clipboard This text will be copied to clipboard" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
+echo "command clipboard This text will be copied to clipboard" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
 
 # Export clipboard to file
-echo "command exportclipboard" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
+echo "command exportclipboard" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
 ....
 
 ===== Application Control
 
 ....
 # Quit Fenrir
-echo "command quitapplication" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
+echo "command quitapplication" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
 ....
 
 ==== Command Reference
@@ -1338,6 +1351,7 @@ echo "command quitapplication" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-dea
 * `+command say <text>+` - Speak the specified text
 * `+command interrupt+` - Stop current speech
 * `+command tempdisablespeech+` - Disable speech until next key press
+* `+ls+` / `+list+` / `+command ls+` / `+command list+` - List registered Fenrir instances
 
 *Clipboard Commands:*
 
@@ -1475,6 +1489,8 @@ By default we ship two sound packs.
 . `+default-wav+` wav encoded, just for compatibility
 
 Sound packs are located at /usr/share/sounds/fenrirscreenreader/
+User sound packs in `+~/.local/stormux/fenrir/sounds/+` are loaded before
+system sound packs.
 
 Sound volume controls how loud the sounds for your selected sound pack
 are.
@@ -1490,38 +1506,38 @@ Values: `+0.0+` is quietest, `+1.0+` is loudest.
 The generic sound driver uses shell commands for play sound and
 frequencies.
 
-`+genericPlayFileCommand+` defines the command that is used to play a
+`+generic_play_file_command+` defines the command that is used to play a
 sound file.
 
 ....
 generic_play_file_command=<your command for playing a file>
 ....
 
-`+genericFrequencyCommand+` defines the command that is used playing
+`+generic_frequency_command+` defines the command that is used playing
 frequencies.
 
 ....
 generic_frequency_command=<your command for playing a frequence>
 ....
 
-The following variables are substituted in `+genericPlayFileCommand+`
-and `+genericFrequencyCommand+`:
+The following variables are substituted in `+generic_play_file_command+`
+and `+generic_frequency_command+`:
 
-* `+fenrirVolume+` = the current volume setting
-* `+fenrirSoundFile+` = the sound file for an sound icon
-* `+fenrirFrequence+` = the frequency to play
-* `+fenrirDuration+` = the duration of the frequency
+* `+fenrir_volume+` = the current volume setting
+* `+fenrir_sound_file+` = the sound file for an sound icon
+* `+fenrir_frequency+` = the frequency to play
+* `+fenrir_duration+` = the duration of the frequency
 
-Example genericPlayFileCommand (default)
+Example generic_play_file_command (default)
 
 ....
-generic_play_file_command=play -q -v fenrirVolume fenrirSoundFile
+generic_play_file_command=play -q -v fenrir_volume fenrir_sound_file
 ....
 
-Example genericFrequencyCommand (default)
+Example generic_frequency_command (default)
 
 ....
-generic_frequency_command=play -q -v fenrirVolume -n -c1 synth fenrirDuration sine fenrirFrequence
+generic_frequency_command=play -q -v fenrir_volume -n -c1 synth fenrir_duration sine fenrir_frequency
 ....
 
 ==== Speech
@@ -1535,8 +1551,14 @@ enabled=True
 Values: on=`+True+`, off=`+False+`
 
 # Select speech driver, options are speechdDriver (default),
-genericDriver or espeakDriver: driver=speechdDriver #driver=espeakDriver
+genericDriver, dectalkDriver, dectalkUsbDriver, litetalkDriver,
+doubletalkDriver or tripletalkDriver: driver=speechdDriver
 #driver=genericDriver
+#driver=dectalkDriver
+#driver=dectalkUsbDriver
+#driver=litetalkDriver
+#driver=doubletalkDriver
+#driver=tripletalkDriver
 
 This Selects the driver used to generate speech output.
 
@@ -1664,13 +1686,39 @@ the pico module:
 language=de-DE
 ....
 
+Serial hardware speech drivers use a serial device. Set an explicit path.
+
+....
+hardware_device=/dev/ttyACM0
+hardware_device=/dev/ttyUSB0
+hardware_device=/dev/ttyS0
+....
+
+Hardware speech drivers use 9600 baud by default.
+
+....
+hardware_baud_rate=9600
+....
+
+The `+doubletalkDriver+` targets DoubleTalk LT-style serial devices. It does
+not support the internal DoubleTalk PC ISA card.
+The `+dectalkUsbDriver+` autodetects the native Access Solutions DECtalk USB,
+USB ID `+0dd0:2001+`. It requires PyUSB and a libusb backend and ignores the
+serial device and baud-rate settings. From a source checkout, install the USB
+dependencies with `+python3 -m pip install '.[dectalk_usb]'+`. A non-root
+Fenrir process needs a local udev rule granting its account read/write access
+to `+0dd0:2001+`; do not make the device world-writable. Other USB hardware
+speech synthesizers are supported only when Linux exposes them as a serial tty
+such as `+/dev/ttyACM0+` or `+/dev/ttyUSB0+`.
+
 Read new text as it occurs auto_read_incoming=True Values: on=`+True+`,
 off=`+False+`
 
 ==== Screen
 
-The settings for screens, (TTY, PTY) are configured in the `+[screen]+`
-section.
+The settings for screen access are configured in the `+[screen]+`
+section. `+vcsaDriver+` is used for Linux TTYs, and `+ptyDriver+` is
+used by X11 terminal mode.
 
 The driver to get the information from the screen:
 
@@ -1780,7 +1828,7 @@ keyboard_layout=desktop
 ....
 
 Values: An absolute Path to a Keyboard definition file or a Filename
-without extension located in `+/etc/fenrir/keyboard+`
+without extension located in `+/etc/fenrirscreenreader/keyboard+`
 
 Announce characters while typing.
 
@@ -1848,7 +1896,7 @@ the current punctuation and dict file in use:
 punctuation_profile=default
 ....
 
-Values: Text, see available profiles in `+/etc/fenrir/punctuation+` or
+Values: Text, see available profiles in `+/etc/fenrirscreenreader/punctuation+` or
 in `+sourceTree/config/punctuation+`
 
 The current punctuation level in use:
@@ -1880,7 +1928,8 @@ link:#export clipboard to file[export clipboard to file]. The variable
 `+$user+` is replaced by the current logged username.
 
 ....
-clipboardExportPath=/tmp/fenrirClipboard
+[clipboard]
+clipboard_export_path=/tmp/fenrirClipboard
 ....
 
 Values: Text, Systemfilepath
@@ -1888,6 +1937,7 @@ Values: Text, Systemfilepath
 The number of available clipboards:
 
 ....
+[clipboard]
 number_of_clipboards=10
 ....
 
@@ -1954,19 +2004,20 @@ Values: Text, see aspell dictionary's.
 Folder Path for your scripts "scriptKey" functionality:
 
 ....
-scriptPath=/usr/share/fenrirscreenreader/scripts
+script_path=/usr/share/fenrirscreenreader/scripts
 ....
 
-Values: Text, Existing path on file system.
+Values: Text, Existing path on file system. User scripts in
+`+~/.local/stormux/fenrir/+` are loaded before this configured path.
 
 Override commands or create new ones without changing the Fenrir
 defaults:
 
 ....
-commandPath=/usr/share/fenrirscreenreader/commands
+command_path=/usr/share/fenrirscreenreader/commands
 ....
 
-Values: Text, Existing path on file system. Subfolders in commandPath
+Values: Text, Existing path on file system. Subfolders in command_path
 are:
 
 * `+commands+` = to create shortcut commands
@@ -2102,14 +2153,15 @@ Values: on=`+True+`, off=`+False+`
 === Scripting
 
 Scripts can be in any language, bash, python, sh or others. Place your
-scripts in the directory /usr/share/fenrirscreenreader/scripts/ (the
-path is configurable in settings.conf). The script key is the
+personal scripts in `+~/.local/stormux/fenrir/+`, or place system scripts in
+the directory /usr/share/fenrirscreenreader/scripts/ (the path is configurable
+in settings.conf). The script key is the
 applications key. Usually this key can be found on the keyboard located
 just left of the right most control key. When you name a script, the key
 name appears in the script seperated by the sequence [.underline]#-#.
 So, for example, if you have a python weather script you want assigned
 to the script key plus the letter w you would name the script
-/usr/share/fenrirscreenreader/scripts/weather[.underline]##-##key_w.py
+~/.local/stormux/fenrir/weather[.underline]##-##key_w.py
 Then, to access the script, simply press the script key and the letter
 w. Scripts must be executable. So, make sure to chmod 755 your script
 when you place it in the scripts directory. The script gets some
@@ -2233,13 +2285,13 @@ that shows the issue.
 
 ==== How-to create a debug file
 
-. Delete old debug stuff +
-`+sudo rm /var/log/fenrir.log+`
 . Start fenrir in debug mode +
 `+sudo fenrir -d+`
 . Do your stuff to reproduce the problem
 . Stop fenrir (`+fenrirKey + q+`)
 
-the debug file is located in `+/var/log/fenrir.log+`
+the debug file is located in `+/tmp/fenrir.log+`. If another Fenrir debug
+instance is already using it, check `+/tmp/fenrir2.log+`,
+`+/tmp/fenrir3.log+`, etc.
 
 Please be as precise as possible to make it easy to solve the problem.

docs/user.md

@@ -38,6 +38,7 @@ Navigate the screen without moving the text cursor. Essential for examining cont
 ### Navigation (Desktop Layout)
 - `Ctrl` - Stop speech (shut up)
 - `Fenrir + H` - Tutorial mode
+- `Fenrir + Ctrl + H` - Speech history
 - `Fenrir + Q` - Quit Fenrir
 - `Fenrir + Keypad 5` - Read current screen
 - `Keypad 8` - Read current line  
@@ -77,7 +78,13 @@ Navigate the screen without moving the text cursor. Essential for examining cont
 ## Configuration
 
 ### Settings File
-Main configuration: `/etc/fenrir/settings/settings.conf`
+Main configuration:
+- **System-wide**: `/etc/fenrirscreenreader/settings/settings.conf`
+- **User-local**: `~/.local/share/stormux/fenrirscreenreader/settings/settings.conf`
+
+When Fenrir runs as a regular user, including `fenrir -x`, it loads the user-local settings file if it exists. If it does not exist, Fenrir falls back to the system settings file, but saving settings creates or updates the user-local file. When Fenrir runs as root, settings save to the system-wide file by default.
+
+User sound themes are loaded before system themes from `~/.local/stormux/fenrir/sounds/`. User scripts are loaded before configured system scripts from `~/.local/stormux/fenrir/`.
 
 ### Key Sections
 - `[speech]` - Speech synthesis settings
@@ -94,6 +101,9 @@ driver=speechdDriver
 rate=0.5
 pitch=0.5
 volume=1.0
+hardware_device=/dev/ttyS0
+hardware_baud_rate=9600
+history_size=50
 
 [sound] 
 enabled=True
@@ -132,59 +142,73 @@ enable_command_remote=True       # allow command execution
 
 ### Basic Usage with socat
 
+#### Instance Discovery
+```bash
+# List registered Fenrir instances and their socket paths
+echo "ls" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
+```
+
+In X terminal mode (`fenrir -x`), multiple Fenrir instances can run at the same
+time. Each instance has a private socket at `/tmp/fenrirscreenreader-<pid>.sock`,
+and one instance may also own the standard control socket. Use `ls` or
+`command ls` on the standard socket to find the private socket for a specific
+instance. Commands sent to the standard socket are handled by its owner when
+possible; otherwise they are forwarded to a registered private socket,
+preferring the sender's Fenrir ancestor when one can be found.
+
 #### Speech Control
 ```bash
 # Interrupt current speech
-echo "command interrupt" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
+echo "command interrupt" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
 
 # Speak custom text  
-echo "command say Hello, this is a test" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
+echo "command say Hello, this is a test" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
 
 # Temporarily disable speech
-echo "command tempdisablespeech" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
+echo "command tempdisablespeech" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
 ```
 
 #### Settings Control
 ```bash
 # Enable highlight tracking
-echo "setting set focus#highlight=True" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
+echo "setting set focus#highlight=True" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
 
 # Change speech parameters
-echo "setting set speech#rate=0.8" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
-echo "setting set speech#pitch=0.6" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
-echo "setting set speech#volume=0.9" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
+echo "setting set speech#rate=0.8" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
+echo "setting set speech#pitch=0.6" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
+echo "setting set speech#volume=0.9" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
 
 # Change punctuation level (none/some/most/all)
-echo "setting set general#punctuation_level=all" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
+echo "setting set general#punctuation_level=all" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
 
 # Voice and TTS control
-echo "setting set speech#voice=en-us+f3" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
-echo "setting set speech#module=espeak-ng" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
+echo "setting set speech#voice=en-us+f3" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
+echo "setting set speech#module=espeak-ng" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
 
 # Multiple settings at once
-echo "setting set speech#rate=0.8;sound#volume=0.7;general#punctuation_level=most" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
+echo "setting set speech#rate=0.8;sound#volume=0.7;general#punctuation_level=most" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
 
 # Reset all settings
-echo "setting reset" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
+echo "setting reset" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
 
 # Save settings
-echo "setting save" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
-echo "setting saveas /tmp/my-settings.conf" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
+echo "setting save" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
+echo "setting saveas /tmp/my-settings.conf" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
 ```
 
 #### Clipboard Operations
 ```bash
 # Add text to clipboard
-echo "command clipboard Text to copy" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
+echo "command clipboard Text to copy" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
 
 # Export clipboard to file
-echo "command exportclipboard" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
+echo "command exportclipboard" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
 ```
 
 #### Application Control
 ```bash
 # Quit Fenrir
-echo "command quitapplication" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
+echo "command quitapplication" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
 ```
 
 ### Command Reference
@@ -193,6 +217,7 @@ echo "command quitapplication" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-dea
 - `command say <text>` - Speak text
 - `command interrupt` - Stop speech
 - `command tempdisablespeech` - Disable until next key
+- `ls` / `list` / `command ls` / `command list` - List registered Fenrir instances
 
 **Settings Commands:**
 - `setting set <section>#<key>=<value>` - Change setting
@@ -219,7 +244,7 @@ echo "command quitapplication" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-dea
 #### Bash Helper Function
 ```bash
 fenrir_say() {
-    echo "command say $1" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-deamon.sock
+    echo "command say $1" | socat - UNIX-CLIENT:/tmp/fenrirscreenreader-daemon.sock
 }
 
 # Usage
@@ -232,7 +257,7 @@ import socket
 import os
 
 def send_fenrir_command(command):
-    socket_path = "/tmp/fenrirscreenreader-deamon.sock"
+    socket_path = "/tmp/fenrirscreenreader-daemon.sock"
     if os.path.exists(socket_path):
         sock = socket.socket(socket.AF_UNIX, socket.SOCK_STREAM)
         try:
@@ -287,6 +312,9 @@ Fenrir automatically detects and provides audio feedback for progress indicators
 - **Automatic**: Works with downloads, compilations, installations
 - **Remote control**: Enable via socket commands
 
+Fenrir detects stable progress structures rather than application-specific
+status formats, which change too frequently to support reliably.
+
 ### Spell Checking
 - `Fenrir + S` - Spell check current word
 - `Fenrir + S S` - Add word to dictionary
@@ -311,6 +339,25 @@ Fenrir automatically detects and provides audio feedback for progress indicators
 ### Speech Drivers
 - **speechdDriver** - Speech-dispatcher (recommended)
 - **genericDriver** - Command-line TTS (espeak, etc.)
+- **dectalkDriver** - Serial DECtalk-compatible hardware speech
+- **dectalkUsbDriver** - Native Access Solutions DECtalk USB (`0dd0:2001`)
+- **litetalkDriver** - Serial LiteTalk-compatible hardware speech
+- **doubletalkDriver** - Serial DoubleTalk LT-compatible hardware speech
+- **tripletalkDriver** - Serial TripleTalk-compatible hardware speech
+
+For serial hardware speech, set `speech#hardware_device` to an explicit path.
+RPITalk gadget mode usually appears as `/dev/ttyACM0`; USB serial adapters
+usually appear as `/dev/ttyUSB0`; built-in serial ports may be `/dev/ttyS0`.
+The default baud rate is `9600`. `doubletalkDriver` targets
+DoubleTalk LT-style serial devices, not the internal DoubleTalk PC ISA card.
+The `dectalkUsbDriver` autodetects the native Access Solutions DECtalk USB and
+ignores both serial settings. It requires PyUSB and a libusb backend. From a
+source checkout, install them with
+`python3 -m pip install '.[dectalk_usb]'`. A non-root Fenrir process also needs
+a local udev rule granting its account read/write access to USB ID
+`0dd0:2001`; do not make the device world-writable. USB TripleTalk devices work
+only if Linux exposes them as a serial tty such as `/dev/ttyACM0` or
+`/dev/ttyUSB0`.
 
 ### Sound Drivers  
 - **genericDriver** - Sox-based (default)
@@ -318,7 +365,7 @@ Fenrir automatically detects and provides audio feedback for progress indicators
 
 ### Input Drivers
 - **evdevDriver** - Linux evdev (recommended for Linux)
-- **ptyDriver** - Terminal emulation (cross-platform)
+- **x11Driver** - X11 terminal-scoped input for `fenrir -x`
 
 ### Screen Drivers
 - **vcsaDriver** - Linux VCSA devices (TTY)
@@ -341,11 +388,35 @@ fenrir [OPTIONS]
 - `-o, --options SECTION#SETTING=VALUE;..` - Override settings
 - `-d, --debug` - Enable debug mode
 - `-p, --print` - Print debug to screen
-- `-e, --emulated-pty` - PTY emulation for desktop use
-- `-E, --emulated-evdev` - PTY + evdev emulation
-- `-F, --force-all-screens` - Ignore ignoreScreen setting
+- `-x, --x11` - PTY + X11 keyboard input scoped to the terminal window
+- `--x11-window-id WINDOWID` - X11 window id to use for `--x11` terminal mode
+- `-F, --force-all-screens` - Ignore ignore_screen setting
 - `-i, -I, --ignore-screen SCREEN` - Ignore specific screen(s), can be used multiple times
 
+### X11 Terminal Mode
+
+Use `fenrir -x` to run Fenrir in an X11 terminal as a normal user:
+
+```bash
+fenrir -x
+```
+
+This mode uses PTY screen monitoring and X11 keyboard input. It is intended for GUI terminals such as xterm and VTE-based terminals, while keeping Fenrir key handling scoped to that terminal window. Desktop screen readers keep their global bindings outside the focused terminal.
+
+Fenrir normally detects the target terminal from `WINDOWID` or the active X11 window. If needed, pass a window id explicitly:
+
+```bash
+fenrir -x --x11-window-id 0x123456
+```
+
+X11 terminal mode uses the same keyboard layout files as TTY Fenrir. Supported Fenrir keys include numpad Insert/`KEY_KP0`, CapsLock/`KEY_CAPSLOCK`, Insert/`KEY_INSERT`, and Super/Windows/`KEY_META`. Compose script keys are exposed as `KEY_COMPOSE`.
+
+In this mode Fenrir runs as a regular user. Settings are loaded from `~/.local/share/stormux/fenrirscreenreader/settings/settings.conf` when present, otherwise from `/etc/fenrirscreenreader/settings/settings.conf`; saving settings creates or updates the user-local file. User sound themes and scripts are read first from `~/.local/stormux/fenrir/sounds/` and `~/.local/stormux/fenrir/`.
+
+This mode requires `python-xlib`.
+
+For a dedicated PTY/terminal screen reader, see TDSR: https://github.com/tspivey/tdsr
+
 ## Troubleshooting
 
 ### No Speech
@@ -366,7 +437,7 @@ fenrir [OPTIONS]
 ### Debug Mode
 ```bash
 sudo fenrir -f -d
-# Debug output goes to /var/log/fenrir.log
+# Debug output goes to /tmp/fenrir.log
 ```
 
 ## Getting Help
@@ -379,5 +450,6 @@ sudo fenrir -f -d
 ## See Also
 
 - [README.md](../README.md) - Installation and basic setup
+- [diff_review_mode.md](./diff_review_mode.md) - Diff review workflow and key bindings
 - [settings.conf](../config/settings/settings.conf) - Configuration reference
-- `man fenrir` - Manual page
+- `man fenrir` - Manual page

docs/user.txt

@@ -857,31 +857,35 @@ Values: ''0.0'' is quietest, ''1.0'' is loudest.
 === Generic Driver ===
 The generic sound driver uses shell commands for play sound and frequencies.
 
-''genericPlayFileCommand'' defines the command that is used to play a sound file.
+''generic_play_file_command'' defines the command that is used to play a sound file.
   generic_play_file_command=<your command for playing a file>
-''genericFrequencyCommand'' defines the command that is used playing frequencies.
+''generic_frequency_command'' defines the command that is used playing frequencies.
   generic_frequency_command=<your command for playing a frequence>
  
-The following variables are substituted in ''genericPlayFileCommand'' and ''genericFrequencyCommand'':
-  * ''fenrirVolume'' = the current volume setting
-  * ''fenrirSoundFile'' = the sound file for an sound icon
-  * ''fenrirFrequence'' = the frequency to play
-  * ''fenrirDuration'' = the duration of the frequency
+The following variables are substituted in ''generic_play_file_command'' and ''generic_frequency_command'':
+  * ''fenrir_volume'' = the current volume setting
+  * ''fenrir_sound_file'' = the sound file for an sound icon
+  * ''fenrir_frequency'' = the frequency to play
+  * ''fenrir_duration'' = the duration of the frequency
 
-Example genericPlayFileCommand (default)
-  generic_play_file_command=play -q -v fenrirVolume fenrirSoundFile
-Example genericFrequencyCommand (default)
-  generic_frequency_command=play -q -v fenrirVolume -n -c1 synth fenrirDuration sine fenrirFrequence
+Example generic_play_file_command (default)
+  generic_play_file_command=play -q -v fenrir_volume fenrir_sound_file
+Example generic_frequency_command (default)
+  generic_frequency_command=play -q -v fenrir_volume -n -c1 synth fenrir_duration sine fenrir_frequency
 ==== Speech ====
 Speech is configured in section ''[speech]''.
 Turn speech on or off:
   enabled=True
 Values: on=''True'', off=''False''
 
-# Select speech driver, options are speechdDriver (default), genericDriver or espeakDriver:
+# Select speech driver, options are speechdDriver (default), genericDriver, dectalkDriver, dectalkUsbDriver, litetalkDriver, doubletalkDriver or tripletalkDriver:
 driver=speechdDriver
-#driver=espeakDriver
 #driver=genericDriver
+#driver=dectalkDriver
+#driver=dectalkUsbDriver
+#driver=litetalkDriver
+#driver=doubletalkDriver
+#driver=tripletalkDriver
 
 Select the driver used to generate speech output.
 
@@ -890,7 +894,11 @@ Select the driver used to generate speech output.
 Available Drivers:
   * ''genericDriver'' using the generic driver, for Fenrir <1.5 this is not available
   * ''speechdDriver'' using speech-dispatcher, for Fenrir <1.5 just use ''speechd''
-  * ''espeakDriver'' using the espeak directly, for Fenrir <1.5 just use ''espeak''
+  * ''dectalkDriver'' using DECtalk-compatible serial hardware or RPITalk
+  * ''dectalkUsbDriver'' using the native Access Solutions DECtalk USB
+  * ''litetalkDriver'' using LiteTalk-compatible serial hardware or RPITalk
+  * ''doubletalkDriver'' using DoubleTalk LT-compatible serial hardware
+  * ''tripletalkDriver'' using TripleTalk-compatible serial hardware
 
 The rate selects how fast Fenrir will speak. 
   rate=0.65
@@ -921,6 +929,17 @@ Select the language you want Fenrir to use.
   language=english-us
 Values: Text, see your TTS synths documentation what is available.
 
+Serial hardware speech drivers use a serial device. Set an explicit path.
+  hardware_device=/dev/ttyACM0
+  hardware_device=/dev/ttyUSB0
+  hardware_device=/dev/ttyS0
+
+Hardware speech drivers use 9600 baud by default.
+  hardware_baud_rate=9600
+
+The doubletalkDriver targets DoubleTalk LT-style serial devices. It does not support the internal DoubleTalk PC ISA card.
+The dectalkUsbDriver autodetects the native Access Solutions DECtalk USB, USB ID 0dd0:2001. It requires PyUSB and a libusb backend and ignores the serial device and baud-rate settings. Other USB hardware speech synthesizers are supported only when Linux exposes them as a serial tty such as /dev/ttyACM0 or /dev/ttyUSB0.
+
 Read new text as it occurs
 auto_read_incoming=True
 Values: on=''True'', off=''False''
@@ -928,22 +947,22 @@ Values: on=''True'', off=''False''
 === Generic Driver ===
 The generic speech driver uses shell commands for speech synthisus.
 
-''genericSpeechCommand'' defines the command that is executed for creating speech
-The following variables are substituted in ''genericSpeechCommand'':
-  * ''FenrirText'' = is the text that should be spoken
-  * ''fenrirModule'' = may be the speech module like used in speech-dispatcher, not every TTY needs this
-  * ''fenrirLanguage'' = the language to speak in
-  * ''fenrirVoice'' = is the current voice that should be used
-  * ''fenrirVolume'' = is replaced with the current volume
-  * ''fenrirPitch'' = is replaced with the current pitch
-  * ''fenrirRate'' = is replaced with the current speed (speech rate)
+''generic_speech_command'' defines the command that is executed for creating speech
+The following variables are substituted in ''generic_speech_command'':
+  * ''fenrir_text'' = is the text that should be spoken
+  * ''fenrir_module'' = may be the speech module like used in speech-dispatcher, not every TTY needs this
+  * ''fenrir_language'' = the language to speak in
+  * ''fenrir_voice'' = is the current voice that should be used
+  * ''fenrir_volume'' = is replaced with the current volume
+  * ''fenrir_pitch'' = is replaced with the current pitch
+  * ''fenrir_rate'' = is replaced with the current speed (speech rate)
 
-Example genericSpeechCommand (default):
-  generic_speech_command=espeak -a fenrirVolume -s fenrirRate -p fenrirPitch -v fenrirVoice "fenrirText"
+Example generic_speech_command (default):
+  generic_speech_command=espeak -a fenrir_volume -s fenrir_rate -p fenrir_pitch -v fenrir_voice "fenrir_text"
 
-These are the minimum and maximum values of the TTS system used in genericSpeechCommand. They are needed to calculate the abstract range in volume, rate and pitch 0.0 - 1.0.
+These are the minimum and maximum values of the TTS system used in generic_speech_command. They are needed to calculate the abstract range in volume, rate and pitch 0.0 - 1.0.
 
-  FenrirMinVolume=0
+  fenrir_min_volume=0
   fenrir_max_volume=200
   fenrir_min_pitch=0
   fenrir_max_pitch=99
@@ -1013,7 +1032,8 @@ Values: Integer,
   * ''0'' = display size
   * ''>0'' number of cells
 ==== Screen ====
-The settings for screens, (TTY, PTY) are configured in the ''[screen]'' section.
+The settings for screen access are configured in the ''[screen]'' section.
+''vcsaDriver'' is used for Linux TTYs, and ''ptyDriver'' is used by X11 terminal mode.
 
 The driver to get the information from the screen:
   driver=vcsaDriver
@@ -1115,10 +1135,12 @@ Values: on=''True'', off=''False''
 Specify the path where the clipboard should be exported to.
 See [[#export clipboard to file|export clipboard to file]].
 The variable ''$user'' is replaced by the current logged username.
-  clipboardExportPath=/tmp/fenrirClipboard
+  [clipboard]
+  clipboard_export_path=/tmp/fenrirClipboard
 Values: Text, Systemfilepath
 
 The number of available clipboards:
+  [clipboard]
   number_of_clipboards=10
 Values: Integer, 1 - 999
 
@@ -1297,10 +1319,11 @@ Please report Bugs and feature requests to:
 
 for bugs please provide a [[#Howto create a debug file|debug]] file that shows the issue.
 ==== How-to create a debug file ====
-  - Delete old debug stuff\\  ''sudo rm /var/log/fenrir.log''
   - Start fenrir in debug mode\\ ''sudo fenrir -d''
   - Do your stuff to reproduce the problem
   - Stop fenrir (''fenrirKey + q'')
-the debug file is located in ''/var/log/fenrir.log''
+the debug file is located in ''/tmp/fenrir.log''. If another Fenrir debug
+instance is already using it, check ''/tmp/fenrir2.log'',
+''/tmp/fenrir3.log'', etc.
 
 Please be as precise as possible to make it easy to solve the problem.

locale/ru/LC_MESSAGES/fenrir.po

@@ -744,7 +744,6 @@ msgid "Script file is not executable"
 msgstr "Файл скрипта не исполняемый"
 
 #: ../src/fenrirscreenreader/commands/commands/temp_disable_speech.py:17
-#: ../src/fenrirscreenreader/commands/onByteInput/15000-enable_temp_speech.py:17
 #: ../src/fenrirscreenreader/commands/onKeyInput/15000-enable_temp_speech.py:17
 msgid "disables speech until next keypress"
 msgstr "Отключить речь пока не нажата следующая клавиша"
@@ -898,7 +897,6 @@ msgid "speech disabled"
 msgstr "Речь выключена"
 
 #: ../src/fenrirscreenreader/commands/commands/toggle_speech.py:25
-#: ../src/fenrirscreenreader/commands/onByteInput/15000-enable_temp_speech.py:24
 #: ../src/fenrirscreenreader/commands/onKeyInput/15000-enable_temp_speech.py:28
 msgid "speech enabled"
 msgstr "Речь включена"
@@ -1089,15 +1087,6 @@ msgstr "Мерцание"
 msgid "default"
 msgstr "По умолчанию"
 
-#: ../src/fenrirscreenreader/core/byteManager.py:103
-#: ../src/fenrirscreenreader/core/byteManager.py:105
-msgid "Sticky Mode On"
-msgstr "Режим залипания включен"
-
-#: ../src/fenrirscreenreader/core/byteManager.py:109
-msgid "bypass"
-msgstr ""
-
 #: ../src/fenrirscreenreader/core/fenrirManager.py:26
 msgid "Start Fenrir"
 msgstr "fenrir запущен"
@@ -1216,4 +1205,3 @@ msgstr "Меню"
 #: ../src/fenrirscreenreader/core/vmenuManager.py:234
 msgid "Action"
 msgstr "Действие"
-

requirements.txt

@@ -1,7 +1,6 @@
 daemonize
 dbus-python
 evdev
-pexpect
 pyenchant
 pyperclip
 pyte

setup.py

@@ -103,9 +103,12 @@ setup(
         "pyudev>=0.21.0",
         "setuptools",
         "setproctitle",
-        "pexpect",
         "pyte>=0.7.0",
     ],
+    extras_require={
+        "dectalk_usb": ["pyusb>=1.2.1"],
+        "x11": ["python-xlib>=0.33"],
+    },
 )
 
 if not forceSettingsFlag:

src/fenrir

@@ -41,7 +41,7 @@ def create_argument_parser():
     argumentParser.add_argument(
         '-s', '--setting',
         metavar='SETTING-FILE',
-        default='/etc/fenrir/settings/settings.conf',
+        default=None,
         help='Path to custom settings file'
     )
     argumentParser.add_argument(
@@ -61,14 +61,15 @@ def create_argument_parser():
         help='Print debug messages to screen'
     )
     argumentParser.add_argument(
-        '-e', '--emulated-pty',
+        '-x', '--x11',
         action='store_true',
-        help='Use PTY emulation with escape sequences for input (enables desktop/X/Wayland usage)'
+        help='Use PTY emulation with X11 keyboard input scoped to the terminal window'
     )
     argumentParser.add_argument(
-        '-E', '--emulated-evdev',
-        action='store_true',
-        help='Use PTY emulation with evdev for input (single instance)'
+        '--x11-window-id',
+        metavar='WINDOWID',
+        default='',
+        help='X11 window id to use for --x11 terminal mode'
     )
     argumentParser.add_argument(
         '-F',
@@ -91,8 +92,8 @@ def validate_arguments(cliArgs):
             if option and ('#' not in option or '=' not in option):
                 return False, f"Invalid option format: {option}\nExpected format: SECTION#SETTING=VALUE"
 
-    if cliArgs.emulated_pty and cliArgs.emulated_evdev:
-        return False, "Cannot use both --emulated-pty and --emulated-evdev simultaneously"
+    if cliArgs.x11_window_id and not cliArgs.x11:
+        return False, "--x11-window-id requires --x11"
 
     return True, None
 
@@ -105,14 +106,17 @@ def run_fenrir():
         fenrirApp.proceed()
     except Exception as e:
         print(f"Error starting Fenrir: {e}", file=sys.stderr)
+        sys.exit(1)
+    except KeyboardInterrupt:
+        print("Interrupted", file=sys.stderr)
+        sys.exit(1)
+    finally:
         if fenrirApp and hasattr(fenrirApp, 'cleanup_on_error'):
             try:
                 fenrirApp.cleanup_on_error()
             except Exception as cleanup_error:
                 print(
                     f"Error during cleanup: {cleanup_error}", file=sys.stderr)
-        sys.exit(1)
-    finally:
         if fenrirApp:
             del fenrirApp
         # Clean up PID file if it exists
@@ -124,6 +128,10 @@ def run_fenrir():
                 pass
 
 
+def should_run_foreground(cliArgs):
+    return cliArgs.foreground or cliArgs.x11
+
+
 def main():
     global cliArgs
     argumentParser = create_argument_parser()
@@ -135,7 +143,7 @@ def main():
         argumentParser.error(errorMsg)
         sys.exit(1)
 
-    if cliArgs.foreground or cliArgs.emulated_pty:
+    if should_run_foreground(cliArgs):
         # Run directly in foreground
         run_fenrir()
     else:

src/fenrirscreenreader/commands/commands/00_init_commands.py

@@ -13,6 +13,8 @@ from fenrirscreenreader.core.i18n import _
 
 
 class command:
+    help_visible = False
+
     def __init__(self):
         pass
 
@@ -22,7 +24,7 @@ class command:
         self.env["runtime"]["MemoryManager"].add_index_list(
             "clipboardHistory",
             self.env["runtime"]["SettingsManager"].get_setting_as_int(
-                "general", "number_of_clipboards"
+                "clipboard", "number_of_clipboards"
             ),
         )
 

src/fenrirscreenreader/commands/commands/apply_tested_voice.py

@@ -34,6 +34,7 @@ class command:
 
             module = self.env["commandBuffer"]["lastTestedModule"]
             voice = self.env["commandBuffer"]["lastTestedVoice"]
+            language = self.env["commandBuffer"].get("lastTestedLanguage")
 
             self.env["runtime"]["OutputManager"].present_text(
                 f"Applying {voice} from {module}", interrupt=True
@@ -46,6 +47,7 @@ class command:
             old_driver = SettingsManager.get_setting("speech", "driver")
             old_module = SettingsManager.get_setting("speech", "module")
             old_voice = SettingsManager.get_setting("speech", "voice")
+            old_language = SettingsManager.get_setting("speech", "language")
 
             try:
                 # Apply new settings to runtime only (use set_setting to update
@@ -55,6 +57,10 @@ class command:
                 )
                 SettingsManager.set_setting("speech", "module", module)
                 SettingsManager.set_setting("speech", "voice", voice)
+                if language:
+                    SettingsManager.set_setting(
+                        "speech", "language", language
+                    )
 
                 # Apply to speech driver instance directly
                 if "SpeechDriver" in self.env["runtime"]:
@@ -62,6 +68,8 @@ class command:
 
                     # Set the module and voice on the driver instance
                     SpeechDriver.set_module(module)
+                    if language:
+                        SpeechDriver.set_language(language)
                     SpeechDriver.set_voice(voice)
 
                     self.env["runtime"]["OutputManager"].present_text(
@@ -77,6 +85,9 @@ class command:
                 SettingsManager.set_setting("speech", "driver", old_driver)
                 SettingsManager.set_setting("speech", "module", old_module)
                 SettingsManager.set_setting("speech", "voice", old_voice)
+                SettingsManager.set_setting(
+                    "speech", "language", old_language
+                )
 
                 self.env["runtime"]["OutputManager"].present_text(
                     f"Failed to apply voice, reverted: {str(e)}",

src/fenrirscreenreader/commands/commands/cycle_keyboard_layout.py

@@ -1,105 +0,0 @@
-#!/usr/bin/env python3
-# -*- coding: utf-8 -*-
-
-# Fenrir TTY screen reader
-# By Chrys, Storm Dragon, and contributors.
-
-import os
-
-from fenrirscreenreader.core import debug
-from fenrirscreenreader.core.i18n import _
-
-
-class command:
-    def __init__(self):
-        pass
-
-    def initialize(self, environment):
-        self.env = environment
-
-    def shutdown(self):
-        pass
-
-    def get_description(self):
-        return _("cycles between available keyboard layouts")
-
-    def get_available_layouts(self):
-        """Get list of available keyboard layout files"""
-        layouts = []
-
-        # Check standard locations for keyboard layouts
-        settings_root = "/etc/fenrirscreenreader/"
-        if not os.path.exists(settings_root):
-            # Fallback to source directory
-            import fenrirscreenreader
-
-            fenrir_path = os.path.dirname(fenrirscreenreader.__file__)
-            settings_root = fenrir_path + "/../../config/"
-
-        keyboard_path = settings_root + "keyboard/"
-
-        if os.path.exists(keyboard_path):
-            for file in os.listdir(keyboard_path):
-                if (
-                    file.endswith(".conf")
-                    and not file.startswith("__")
-                    and not file.lower().startswith("pty")
-                ):
-                    layout_name = file.replace(".conf", "")
-                    if layout_name not in layouts:
-                        layouts.append(layout_name)
-
-        # Ensure we have at least basic layouts
-        if not layouts:
-            layouts = ["desktop", "laptop"]
-        else:
-            layouts.sort()
-
-        return layouts
-
-    def run(self):
-        current_layout = self.env["runtime"]["SettingsManager"].get_setting(
-            "keyboard", "keyboard_layout"
-        )
-
-        # Extract layout name from full path if needed
-        if "/" in current_layout:
-            current_layout = os.path.basename(current_layout).replace(
-                ".conf", ""
-            )
-
-        # Get available layouts
-        available_layouts = self.get_available_layouts()
-
-        # Find next layout in cycle
-        try:
-            current_index = available_layouts.index(current_layout)
-            next_index = (current_index + 1) % len(available_layouts)
-        except ValueError:
-            # If current layout not found, start from beginning
-            next_index = 0
-
-        next_layout = available_layouts[next_index]
-
-        # Update setting and reload shortcuts
-        self.env["runtime"]["SettingsManager"].set_setting(
-            "keyboard", "keyboard_layout", next_layout
-        )
-
-        # Reload shortcuts with new layout
-        try:
-            self.env["runtime"]["InputManager"].reload_shortcuts()
-            self.env["runtime"]["OutputManager"].present_text(
-                _("Switched to {} keyboard layout").format(next_layout),
-                interrupt=True,
-            )
-        except Exception as e:
-            self.env["runtime"]["DebugManager"].write_debug_out(
-                "Error reloading shortcuts: " + str(e), debug.DebugLevel.ERROR
-            )
-            self.env["runtime"]["OutputManager"].present_text(
-                _("Error switching keyboard layout"), interrupt=True
-            )
-
-    def set_callback(self, callback):
-        pass

src/fenrirscreenreader/commands/commands/diff_key_help.py

@@ -0,0 +1,27 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+
+# Fenrir TTY screen reader
+# By Chrys, Storm Dragon, and contributors.
+
+from fenrirscreenreader.core.i18n import _
+
+
+class command:
+    def __init__(self):
+        pass
+
+    def initialize(self, environment):
+        self.env = environment
+
+    def shutdown(self):
+        pass
+
+    def get_description(self):
+        return _("speak diff review key help")
+
+    def run(self):
+        self.env["runtime"]["DiffReviewManager"].present_key_help()
+
+    def set_callback(self, callback):
+        pass

src/fenrirscreenreader/commands/commands/diff_next_added.py

@@ -0,0 +1,27 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+
+# Fenrir TTY screen reader
+# By Chrys, Storm Dragon, and contributors.
+
+from fenrirscreenreader.core.i18n import _
+
+
+class command:
+    def __init__(self):
+        pass
+
+    def initialize(self, environment):
+        self.env = environment
+
+    def shutdown(self):
+        pass
+
+    def get_description(self):
+        return _("move to next added diff line")
+
+    def run(self):
+        self.env["runtime"]["DiffReviewManager"].next_added()
+
+    def set_callback(self, callback):
+        pass

src/fenrirscreenreader/commands/commands/diff_next_char.py

@@ -0,0 +1,27 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+
+# Fenrir TTY screen reader
+# By Chrys, Storm Dragon, and contributors.
+
+from fenrirscreenreader.core.i18n import _
+
+
+class command:
+    def __init__(self):
+        pass
+
+    def initialize(self, environment):
+        self.env = environment
+
+    def shutdown(self):
+        pass
+
+    def get_description(self):
+        return _("move to next character in diff line")
+
+    def run(self):
+        self.env["runtime"]["DiffReviewManager"].next_char()
+
+    def set_callback(self, callback):
+        pass

src/fenrirscreenreader/commands/commands/diff_next_file_section.py

@@ -0,0 +1,27 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+
+# Fenrir TTY screen reader
+# By Chrys, Storm Dragon, and contributors.
+
+from fenrirscreenreader.core.i18n import _
+
+
+class command:
+    def __init__(self):
+        pass
+
+    def initialize(self, environment):
+        self.env = environment
+
+    def shutdown(self):
+        pass
+
+    def get_description(self):
+        return _("move to next diff file section")
+
+    def run(self):
+        self.env["runtime"]["DiffReviewManager"].next_file_section()
+
+    def set_callback(self, callback):
+        pass

src/fenrirscreenreader/commands/commands/diff_next_hunk.py

@@ -0,0 +1,27 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+
+# Fenrir TTY screen reader
+# By Chrys, Storm Dragon, and contributors.
+
+from fenrirscreenreader.core.i18n import _
+
+
+class command:
+    def __init__(self):
+        pass
+
+    def initialize(self, environment):
+        self.env = environment
+
+    def shutdown(self):
+        pass
+
+    def get_description(self):
+        return _("move to next diff hunk")
+
+    def run(self):
+        self.env["runtime"]["DiffReviewManager"].next_hunk()
+
+    def set_callback(self, callback):
+        pass

src/fenrirscreenreader/commands/commands/diff_next_line.py

@@ -0,0 +1,27 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+
+# Fenrir TTY screen reader
+# By Chrys, Storm Dragon, and contributors.
+
+from fenrirscreenreader.core.i18n import _
+
+
+class command:
+    def __init__(self):
+        pass
+
+    def initialize(self, environment):
+        self.env = environment
+
+    def shutdown(self):
+        pass
+
+    def get_description(self):
+        return _("move to next diff line")
+
+    def run(self):
+        self.env["runtime"]["DiffReviewManager"].next_line()
+
+    def set_callback(self, callback):
+        pass

src/fenrirscreenreader/commands/commands/diff_next_removed.py

@@ -0,0 +1,27 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+
+# Fenrir TTY screen reader
+# By Chrys, Storm Dragon, and contributors.
+
+from fenrirscreenreader.core.i18n import _
+
+
+class command:
+    def __init__(self):
+        pass
+
+    def initialize(self, environment):
+        self.env = environment
+
+    def shutdown(self):
+        pass
+
+    def get_description(self):
+        return _("move to next removed diff line")
+
+    def run(self):
+        self.env["runtime"]["DiffReviewManager"].next_removed()
+
+    def set_callback(self, callback):
+        pass

src/fenrirscreenreader/commands/commands/diff_next_word.py

@@ -0,0 +1,27 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+
+# Fenrir TTY screen reader
+# By Chrys, Storm Dragon, and contributors.
+
+from fenrirscreenreader.core.i18n import _
+
+
+class command:
+    def __init__(self):
+        pass
+
+    def initialize(self, environment):
+        self.env = environment
+
+    def shutdown(self):
+        pass
+
+    def get_description(self):
+        return _("move to next word in diff line")
+
+    def run(self):
+        self.env["runtime"]["DiffReviewManager"].next_word()
+
+    def set_callback(self, callback):
+        pass

src/fenrirscreenreader/commands/commands/diff_prev_added.py

@@ -0,0 +1,27 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+
+# Fenrir TTY screen reader
+# By Chrys, Storm Dragon, and contributors.
+
+from fenrirscreenreader.core.i18n import _
+
+
+class command:
+    def __init__(self):
+        pass
+
+    def initialize(self, environment):
+        self.env = environment
+
+    def shutdown(self):
+        pass
+
+    def get_description(self):
+        return _("move to previous added diff line")
+
+    def run(self):
+        self.env["runtime"]["DiffReviewManager"].prev_added()
+
+    def set_callback(self, callback):
+        pass

src/fenrirscreenreader/commands/commands/diff_prev_char.py

@@ -0,0 +1,27 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+
+# Fenrir TTY screen reader
+# By Chrys, Storm Dragon, and contributors.
+
+from fenrirscreenreader.core.i18n import _
+
+
+class command:
+    def __init__(self):
+        pass
+
+    def initialize(self, environment):
+        self.env = environment
+
+    def shutdown(self):
+        pass
+
+    def get_description(self):
+        return _("move to previous character in diff line")
+
+    def run(self):
+        self.env["runtime"]["DiffReviewManager"].prev_char()
+
+    def set_callback(self, callback):
+        pass

src/fenrirscreenreader/commands/commands/diff_prev_file_section.py

@@ -0,0 +1,27 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+
+# Fenrir TTY screen reader
+# By Chrys, Storm Dragon, and contributors.
+
+from fenrirscreenreader.core.i18n import _
+
+
+class command:
+    def __init__(self):
+        pass
+
+    def initialize(self, environment):
+        self.env = environment
+
+    def shutdown(self):
+        pass
+
+    def get_description(self):
+        return _("move to previous diff file section")
+
+    def run(self):
+        self.env["runtime"]["DiffReviewManager"].prev_file_section()
+
+    def set_callback(self, callback):
+        pass

src/fenrirscreenreader/commands/commands/diff_prev_hunk.py

@@ -0,0 +1,27 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+
+# Fenrir TTY screen reader
+# By Chrys, Storm Dragon, and contributors.
+
+from fenrirscreenreader.core.i18n import _
+
+
+class command:
+    def __init__(self):
+        pass
+
+    def initialize(self, environment):
+        self.env = environment
+
+    def shutdown(self):
+        pass
+
+    def get_description(self):
+        return _("move to previous diff hunk")
+
+    def run(self):
+        self.env["runtime"]["DiffReviewManager"].prev_hunk()
+
+    def set_callback(self, callback):
+        pass

src/fenrirscreenreader/commands/commands/diff_prev_line.py

@@ -0,0 +1,27 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+
+# Fenrir TTY screen reader
+# By Chrys, Storm Dragon, and contributors.
+
+from fenrirscreenreader.core.i18n import _
+
+
+class command:
+    def __init__(self):
+        pass
+
+    def initialize(self, environment):
+        self.env = environment
+
+    def shutdown(self):
+        pass
+
+    def get_description(self):
+        return _("move to previous diff line")
+
+    def run(self):
+        self.env["runtime"]["DiffReviewManager"].prev_line()
+
+    def set_callback(self, callback):
+        pass

src/fenrirscreenreader/commands/commands/diff_prev_removed.py

@@ -0,0 +1,27 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+
+# Fenrir TTY screen reader
+# By Chrys, Storm Dragon, and contributors.
+
+from fenrirscreenreader.core.i18n import _
+
+
+class command:
+    def __init__(self):
+        pass
+
+    def initialize(self, environment):
+        self.env = environment
+
+    def shutdown(self):
+        pass
+
+    def get_description(self):
+        return _("move to previous removed diff line")
+
+    def run(self):
+        self.env["runtime"]["DiffReviewManager"].prev_removed()
+
+    def set_callback(self, callback):
+        pass

src/fenrirscreenreader/commands/commands/diff_prev_word.py

@@ -0,0 +1,27 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+
+# Fenrir TTY screen reader
+# By Chrys, Storm Dragon, and contributors.
+
+from fenrirscreenreader.core.i18n import _
+
+
+class command:
+    def __init__(self):
+        pass
+
+    def initialize(self, environment):
+        self.env = environment
+
+    def shutdown(self):
+        pass
+
+    def get_description(self):
+        return _("move to previous word in diff line")
+
+    def run(self):
+        self.env["runtime"]["DiffReviewManager"].prev_word()
+
+    def set_callback(self, callback):
+        pass

src/fenrirscreenreader/commands/commands/diff_summary.py

@@ -0,0 +1,27 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+
+# Fenrir TTY screen reader
+# By Chrys, Storm Dragon, and contributors.
+
+from fenrirscreenreader.core.i18n import _
+
+
+class command:
+    def __init__(self):
+        pass
+
+    def initialize(self, environment):
+        self.env = environment
+
+    def shutdown(self):
+        pass
+
+    def get_description(self):
+        return _("speak diff summary")
+
+    def run(self):
+        self.env["runtime"]["DiffReviewManager"].present_summary()
+
+    def set_callback(self, callback):
+        pass

src/fenrirscreenreader/commands/commands/exit_diff_mode.py

@@ -0,0 +1,32 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+
+# Fenrir TTY screen reader
+# By Chrys, Storm Dragon, and contributors.
+
+from fenrirscreenreader.core.i18n import _
+
+
+class command:
+    def __init__(self):
+        pass
+
+    def initialize(self, environment):
+        self.env = environment
+
+    def shutdown(self):
+        pass
+
+    def get_description(self):
+        return _("exit diff review mode")
+
+    def run(self):
+        if not self.env["runtime"]["DiffReviewManager"].is_active():
+            return
+        self.env["runtime"]["DiffReviewManager"].disable_mode()
+        self.env["runtime"]["DiffReviewManager"].present_mode_message(
+            _("Diff review mode disabled."),
+        )
+
+    def set_callback(self, callback):
+        pass

src/fenrirscreenreader/commands/commands/export_clipboard_to_file.py

@@ -26,7 +26,7 @@ class command:
     def run(self):
         clipboard_file_path = self.env["runtime"][
             "SettingsManager"
-        ].get_setting("general", "clipboard_export_path")
+        ].get_setting("clipboard", "clipboard_export_path")
         clipboard_file_path = clipboard_file_path.replace(
             "$user", self.env["general"]["curr_user"]
         )

src/fenrirscreenreader/commands/commands/export_clipboard_to_x.py

@@ -4,33 +4,35 @@
 # Fenrir TTY screen reader
 # By Chrys, Storm Dragon, and contributors.
 
-import _thread
-import importlib
-import os
-
-import pyperclip
-
+from fenrirscreenreader.core import debug
 from fenrirscreenreader.core.i18n import _
+from fenrirscreenreader.utils import x_clipboard
+
+
+def write_clipboard(clipboard):
+    """Write the graphical clipboard without accessing Fenrir state."""
+    return x_clipboard.write_text(clipboard, scan_displays=True)
 
 
 class command:
     def __init__(self):
-        pass
+        self._task_ids = set()
 
     def initialize(self, environment, script_path=""):
         self.env = environment
         self.script_path = script_path
 
     def shutdown(self):
-        pass
+        task_manager = self.env["runtime"].get("BackgroundTaskManager")
+        if task_manager:
+            for task_id in self._task_ids:
+                task_manager.cancel_task(task_id)
+        self._task_ids.clear()
 
     def get_description(self):
         return _("Export current fenrir clipboard to X or GUI clipboard")
 
     def run(self):
-        _thread.start_new_thread(self._thread_run, ())
-
-    def _thread_run(self):
         try:
             # Check if clipboard is empty
             if self.env["runtime"]["MemoryManager"].is_index_list_empty(
@@ -46,52 +48,43 @@ class command:
                 "MemoryManager"
             ].get_index_list_element("clipboardHistory")
 
-            # Remember original display environment variable if it exists
-            original_display = os.environ.get("DISPLAY", "")
-            success = False
-
-            # Try different display options
-            for i in range(10):
-                display = f":{i}"
-                try:
-                    # Set display environment variable
-                    os.environ["DISPLAY"] = display
-                    # Attempt to set clipboard content
-                    # Weird workaround for some distros
-                    importlib.reload(pyperclip)
-                    pyperclip.copy(clipboard)
-                    # If we get here without exception, we found a working
-                    # display
-                    success = True
-                    break
-                except Exception:
-                    # Failed for this display, try next one
-                    continue
-
-            # Restore original display setting
-            if original_display:
-                os.environ["DISPLAY"] = original_display
-            else:
-                os.environ.pop("DISPLAY", None)
-
-            # Notify the user of the result
-            if success:
-                self.env["runtime"]["OutputManager"].present_text(
-                    _("exported to the X session."), interrupt=True
-                )
-            else:
-                self.env["runtime"]["OutputManager"].present_text(
-                    _(
-                        "failed to export to X clipboard. No available display "
-                        "found."
-                    ),
-                    interrupt=True,
-                )
+            task_id = self.env["runtime"][
+                "BackgroundTaskManager"
+            ].submit_task(
+                write_clipboard,
+                lambda result: self._handle_result(clipboard, result),
+                clipboard,
+            )
+            if task_id is not None:
+                self._task_ids.add(task_id)
 
         except Exception as e:
             self.env["runtime"]["OutputManager"].present_text(
                 str(e), sound_icon="", interrupt=False
             )
 
+    def _handle_result(self, clipboard, result):
+        self._task_ids.discard(result.get("task_id"))
+        if result.get("succeeded") and result.get("value"):
+            sync_manager = self.env["runtime"].get("ClipboardSyncManager")
+            if sync_manager:
+                sync_manager.mark_written_to_x(clipboard)
+            self.env["runtime"]["OutputManager"].present_text(
+                _("exported to the X session."), interrupt=True
+            )
+            return
+
+        if not result.get("succeeded") and result.get("error"):
+            self.env["runtime"]["DebugManager"].write_debug_out(
+                "Clipboard export failed: " + result["error"],
+                debug.DebugLevel.ERROR,
+            )
+        self.env["runtime"]["OutputManager"].present_text(
+            _(
+                "failed to export to X clipboard. No available display found."
+            ),
+            interrupt=True,
+        )
+
     def set_callback(self, callback):
         pass

src/fenrirscreenreader/commands/commands/forward_keypress.py

@@ -22,7 +22,10 @@ class command:
         return _("sends the following keypress to the terminal or application")
 
     def run(self):
-        self.env["input"]["key_forward"] = 3
+        if self.env["runtime"]["InputManager"].no_key_pressed():
+            self.env["input"]["key_forward"] = 1
+        else:
+            self.env["input"]["key_forward"] = -1
         self.env["runtime"]["OutputManager"].present_text(
             _("Forward next keypress"), interrupt=True
         )

src/fenrirscreenreader/commands/commands/import_clipboard_from_file.py

@@ -26,7 +26,7 @@ class command:
     def run(self):
         clipboard_file_path = self.env["runtime"][
             "SettingsManager"
-        ].get_setting("general", "clipboard_export_path")
+        ].get_setting("clipboard", "clipboard_export_path")
         clipboard_file_path = clipboard_file_path.replace(
             "$user", self.env["general"]["curr_user"]
         )

src/fenrirscreenreader/commands/commands/import_clipboard_from_x.py

@@ -4,84 +4,68 @@
 # Fenrir TTY screen reader
 # By Chrys, Storm Dragon, and contributors.
 
-import _thread
-import importlib
-import os
-
-import pyperclip
-
 from fenrirscreenreader.core.i18n import _
+from fenrirscreenreader.utils import x_clipboard
+
+
+def read_clipboard():
+    """Read the graphical clipboard without accessing Fenrir state."""
+    return x_clipboard.read_text(scan_displays=True)
 
 
 class command:
     def __init__(self):
-        pass
+        self._task_ids = set()
 
     def initialize(self, environment, script_path=""):
         self.env = environment
         self.script_path = script_path
 
     def shutdown(self):
-        pass
+        task_manager = self.env["runtime"].get("BackgroundTaskManager")
+        if task_manager:
+            for task_id in self._task_ids:
+                task_manager.cancel_task(task_id)
+        self._task_ids.clear()
 
     def get_description(self):
         return _("imports the graphical clipboard to Fenrir's clipboard")
 
     def run(self):
-        _thread.start_new_thread(self._thread_run, ())
+        task_id = self.env["runtime"]["BackgroundTaskManager"].submit_task(
+            read_clipboard, self._handle_result
+        )
+        if task_id is not None:
+            self._task_ids.add(task_id)
 
-    def _thread_run(self):
-        try:
-            # Remember original display environment variable if it exists
-            original_display = os.environ.get("DISPLAY", "")
-            clipboard_content = None
-
-            # Try different display options
-            for i in range(10):
-                display = f":{i}"
-                try:
-                    # Set display environment variable
-                    os.environ["DISPLAY"] = display
-                    # Attempt to get clipboard content
-                    # Weird workaround for some distros
-                    importlib.reload(pyperclip)
-                    clipboard_content = pyperclip.paste()
-                    # If we get here without exception, we found a working
-                    # display
-                    if clipboard_content:
-                        break
-                except Exception:
-                    # Failed for this display, try next one
-                    continue
-
-            # Restore original display setting
-            if original_display:
-                os.environ["DISPLAY"] = original_display
-            else:
-                os.environ.pop("DISPLAY", None)
-
-            # Process the clipboard content if we found any
-            if clipboard_content and isinstance(clipboard_content, str):
-                self.env["runtime"]["MemoryManager"].add_value_to_first_index(
-                    "clipboardHistory", clipboard_content
-                )
-                self.env["runtime"]["OutputManager"].present_text(
-                    "Import to Clipboard",
-                    sound_icon="CopyToClipboard",
-                    interrupt=True,
-                )
-                self.env["runtime"]["OutputManager"].present_text(
-                    clipboard_content, sound_icon="", interrupt=False
-                )
-            else:
-                self.env["runtime"]["OutputManager"].present_text(
-                    "No text found in clipboard or no accessible display",
-                    interrupt=True,
-                )
-        except Exception as e:
+    def _handle_result(self, result):
+        self._task_ids.discard(result.get("task_id"))
+        clipboard_content = result.get("value")
+        if not result.get("succeeded"):
             self.env["runtime"]["OutputManager"].present_text(
-                str(e), sound_icon="", interrupt=False
+                result.get("error", "Clipboard import failed"),
+                sound_icon="",
+                interrupt=False,
             )
+            return
+        if not clipboard_content or not isinstance(clipboard_content, str):
+            self.env["runtime"]["OutputManager"].present_text(
+                "No text found in clipboard or no accessible display",
+                interrupt=True,
+            )
+            return
+
+        self.env["runtime"]["MemoryManager"].add_value_to_first_index(
+            "clipboardHistory", clipboard_content
+        )
+        self.env["runtime"]["OutputManager"].present_text(
+            "Import to Clipboard",
+            sound_icon="CopyToClipboard",
+            interrupt=True,
+        )
+        self.env["runtime"]["OutputManager"].present_text(
+            clipboard_content, sound_icon="", interrupt=False
+        )
 
     def set_callback(self, callback):
         pass

src/fenrirscreenreader/commands/commands/paste_clipboard.py

@@ -18,7 +18,7 @@ class command:
         self.env["runtime"]["MemoryManager"].add_index_list(
             "clipboardHistory",
             self.env["runtime"]["SettingsManager"].get_setting_as_int(
-                "general", "number_of_clipboards"
+                "clipboard", "number_of_clipboards"
             ),
         )
 

src/fenrirscreenreader/commands/commands/review_curr_word_phonetic.py

@@ -38,7 +38,7 @@ class command:
             self.env["screen"]["new_content_text"],
         )
 
-        if curr_word.isspace():
+        if not curr_word or curr_word.isspace():
             self.env["runtime"]["OutputManager"].present_text(
                 _("blank"), interrupt=True, flush=False
             )

src/fenrirscreenreader/commands/commands/review_curr_word_spell.py

@@ -20,7 +20,7 @@ class command:
         pass
 
     def get_description(self):
-        return _("Phonetically spells the next word and moves review to it")
+        return _("Spells the current word")
 
     def run(self):
         self.env["runtime"][
@@ -29,30 +29,28 @@ class command:
         (
             self.env["screen"]["newCursorReview"]["x"],
             self.env["screen"]["newCursorReview"]["y"],
-            next_word,
+            curr_word,
             end_of_screen,
             line_break,
-        ) = word_utils.get_next_word(
+        ) = word_utils.get_current_word(
             self.env["screen"]["newCursorReview"]["x"],
             self.env["screen"]["newCursorReview"]["y"],
             self.env["screen"]["new_content_text"],
         )
 
-        if next_word.isspace():
+        if not curr_word or curr_word.isspace():
             self.env["runtime"]["OutputManager"].present_text(
                 _("blank"), interrupt=True, flush=False
             )
         else:
-            first_sequence = True
-            for c in next_word:
-                curr_char = char_utils.get_phonetic(c)
-                self.env["runtime"]["OutputManager"].present_text(
+            for index, curr_char in enumerate(curr_word):
+                char_utils.present_char_for_review(
+                    self.env,
                     curr_char,
-                    interrupt=first_sequence,
+                    interrupt=index == 0,
                     announce_capital=True,
                     flush=False,
                 )
-                first_sequence = False
         if end_of_screen:
             if self.env["runtime"]["SettingsManager"].get_setting_as_bool(
                 "review", "end_of_screen"

src/fenrirscreenreader/commands/commands/review_next_char_phonetic.py

@@ -1,65 +0,0 @@
-#!/usr/bin/env python3
-# -*- coding: utf-8 -*-
-
-# Fenrir TTY screen reader
-# By Chrys, Storm Dragon, and contributors.
-
-from fenrirscreenreader.core.i18n import _
-from fenrirscreenreader.utils import char_utils
-
-
-class command:
-    def __init__(self):
-        pass
-
-    def initialize(self, environment):
-        self.env = environment
-
-    def shutdown(self):
-        pass
-
-    def get_description(self):
-        return _(
-            "phonetically presents the next character and set review to it"
-        )
-
-    def run(self):
-        self.env["runtime"][
-            "CursorManager"
-        ].enter_review_mode_curr_text_cursor()
-
-        (
-            self.env["screen"]["newCursorReview"]["x"],
-            self.env["screen"]["newCursorReview"]["y"],
-            next_char,
-            end_of_screen,
-            line_break,
-        ) = char_utils.get_next_char(
-            self.env["screen"]["newCursorReview"]["x"],
-            self.env["screen"]["newCursorReview"]["y"],
-            self.env["screen"]["new_content_text"],
-        )
-
-        next_char = char_utils.get_phonetic(next_char)
-        self.env["runtime"]["OutputManager"].present_text(
-            next_char, interrupt=True, announce_capital=True, flush=False
-        )
-        if end_of_screen:
-            if self.env["runtime"]["SettingsManager"].get_setting_as_bool(
-                "review", "end_of_screen"
-            ):
-                self.env["runtime"]["OutputManager"].present_text(
-                    _("end of screen"),
-                    interrupt=True,
-                    sound_icon="EndOfScreen",
-                )
-        if line_break:
-            if self.env["runtime"]["SettingsManager"].get_setting_as_bool(
-                "review", "line_break"
-            ):
-                self.env["runtime"]["OutputManager"].present_text(
-                    _("line break"), interrupt=False, sound_icon="EndOfLine"
-                )
-
-    def set_callback(self, callback):
-        pass

src/fenrirscreenreader/commands/commands/review_prev_char.py

@@ -112,9 +112,9 @@ class command:
                 "review", "end_of_screen"
             ):
                 self.env["runtime"]["OutputManager"].present_text(
-                    _("end of screen"),
+                    _("start of screen"),
                     interrupt=True,
-                    sound_icon="EndOfScreen",
+                    sound_icon="StartOfScreen",
                 )
         if line_break:
             if self.env["runtime"]["SettingsManager"].get_setting_as_bool(

src/fenrirscreenreader/commands/commands/review_prev_char_phonetic.py

@@ -1,65 +0,0 @@
-#!/usr/bin/env python3
-# -*- coding: utf-8 -*-
-
-# Fenrir TTY screen reader
-# By Chrys, Storm Dragon, and contributors.
-
-from fenrirscreenreader.core.i18n import _
-from fenrirscreenreader.utils import char_utils
-
-
-class command:
-    def __init__(self):
-        pass
-
-    def initialize(self, environment):
-        self.env = environment
-
-    def shutdown(self):
-        pass
-
-    def get_description(self):
-        return _(
-            "phonetically presents the previous character and set review to it"
-        )
-
-    def run(self):
-        self.env["runtime"][
-            "CursorManager"
-        ].enter_review_mode_curr_text_cursor()
-
-        (
-            self.env["screen"]["newCursorReview"]["x"],
-            self.env["screen"]["newCursorReview"]["y"],
-            prev_char,
-            end_of_screen,
-            line_break,
-        ) = char_utils.get_prev_char(
-            self.env["screen"]["newCursorReview"]["x"],
-            self.env["screen"]["newCursorReview"]["y"],
-            self.env["screen"]["new_content_text"],
-        )
-
-        prev_char = char_utils.get_phonetic(prev_char)
-        self.env["runtime"]["OutputManager"].present_text(
-            prev_char, interrupt=True, announce_capital=True, flush=False
-        )
-        if end_of_screen:
-            if self.env["runtime"]["SettingsManager"].get_setting_as_bool(
-                "review", "end_of_screen"
-            ):
-                self.env["runtime"]["OutputManager"].present_text(
-                    _("end of screen"),
-                    interrupt=True,
-                    sound_icon="EndOfScreen",
-                )
-        if line_break:
-            if self.env["runtime"]["SettingsManager"].get_setting_as_bool(
-                "review", "line_break"
-            ):
-                self.env["runtime"]["OutputManager"].present_text(
-                    _("line break"), interrupt=False, sound_icon="EndOfLine"
-                )
-
-    def set_callback(self, callback):
-        pass

src/fenrirscreenreader/commands/commands/review_prev_line.py

@@ -50,9 +50,9 @@ class command:
                 "review", "end_of_screen"
             ):
                 self.env["runtime"]["OutputManager"].present_text(
-                    _("end of screen"),
+                    _("start of screen"),
                     interrupt=True,
-                    sound_icon="EndOfScreen",
+                    sound_icon="StartOfScreen",
                 )
 
     def set_callback(self, callback):

src/fenrirscreenreader/commands/commands/review_prev_word.py

@@ -95,9 +95,9 @@ class command:
                 "review", "end_of_screen"
             ):
                 self.env["runtime"]["OutputManager"].present_text(
-                    _("end of screen"),
+                    _("start of screen"),
                     interrupt=False,
-                    sound_icon="EndOfScreen",
+                    sound_icon="StartOfScreen",
                 )
         if line_break:
             if self.env["runtime"]["SettingsManager"].get_setting_as_bool(

src/fenrirscreenreader/commands/commands/review_prev_word_phonetic.py

@@ -1,76 +0,0 @@
-#!/usr/bin/env python3
-# -*- coding: utf-8 -*-
-
-# Fenrir TTY screen reader
-# By Chrys, Storm Dragon, and contributors.
-
-from fenrirscreenreader.core.i18n import _
-from fenrirscreenreader.utils import char_utils
-from fenrirscreenreader.utils import word_utils
-
-
-class command:
-    def __init__(self):
-        pass
-
-    def initialize(self, environment):
-        self.env = environment
-
-    def shutdown(self):
-        pass
-
-    def get_description(self):
-        return _(
-            "Phonetically spells the previous word and moves review to it"
-        )
-
-    def run(self):
-        self.env["runtime"][
-            "CursorManager"
-        ].enter_review_mode_curr_text_cursor()
-        (
-            self.env["screen"]["newCursorReview"]["x"],
-            self.env["screen"]["newCursorReview"]["y"],
-            prev_word,
-            end_of_screen,
-            line_break,
-        ) = word_utils.get_prev_word(
-            self.env["screen"]["newCursorReview"]["x"],
-            self.env["screen"]["newCursorReview"]["y"],
-            self.env["screen"]["new_content_text"],
-        )
-
-        if prev_word.isspace():
-            self.env["runtime"]["OutputManager"].present_text(
-                _("blank"), interrupt=True, flush=False
-            )
-        else:
-            first_sequence = True
-            for c in prev_word:
-                curr_char = char_utils.get_phonetic(c)
-                self.env["runtime"]["OutputManager"].present_text(
-                    curr_char,
-                    interrupt=first_sequence,
-                    announce_capital=True,
-                    flush=False,
-                )
-                first_sequence = False
-        if end_of_screen:
-            if self.env["runtime"]["SettingsManager"].get_setting_as_bool(
-                "review", "end_of_screen"
-            ):
-                self.env["runtime"]["OutputManager"].present_text(
-                    _("end of screen"),
-                    interrupt=True,
-                    sound_icon="EndOfScreen",
-                )
-        if line_break:
-            if self.env["runtime"]["SettingsManager"].get_setting_as_bool(
-                "review", "line_break"
-            ):
-                self.env["runtime"]["OutputManager"].present_text(
-                    _("line break"), interrupt=False, sound_icon="EndOfLine"
-                )
-
-    def set_callback(self, callback):
-        pass

src/fenrirscreenreader/commands/commands/review_up.py

@@ -50,9 +50,9 @@ class command:
                 "review", "end_of_screen"
             ):
                 self.env["runtime"]["OutputManager"].present_text(
-                    _("end of screen"),
+                    _("start of screen"),
                     interrupt=True,
-                    sound_icon="EndOfScreen",
+                    sound_icon="StartOfScreen",
                 )
         if line_break:
             if self.env["runtime"]["SettingsManager"].get_setting_as_bool(

src/fenrirscreenreader/commands/commands/save_settings.py

@@ -22,10 +22,7 @@ class command:
         return _("Saves your current Fenrir settings so they are the default.")
 
     def run(self):
-        settings_file = self.env["runtime"][
-            "SettingsManager"
-        ].get_settings_file()
-        self.env["runtime"]["SettingsManager"].save_settings(settings_file)
+        self.env["runtime"]["SettingsManager"].save_settings()
         self.env["runtime"]["OutputManager"].present_text(
             _("Settings saved."), interrupt=True
         )

src/fenrirscreenreader/commands/commands/speech_history.py

@@ -0,0 +1,27 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+
+# Fenrir TTY screen reader
+# By Chrys, Storm Dragon, and contributors.
+
+from fenrirscreenreader.core.i18n import _
+
+
+class command:
+    def __init__(self):
+        pass
+
+    def initialize(self, environment):
+        self.env = environment
+
+    def shutdown(self):
+        pass
+
+    def get_description(self):
+        return _("opens speech history")
+
+    def run(self):
+        self.env["runtime"]["SpeechHistoryManager"].open_history()
+
+    def set_callback(self, callback):
+        pass

src/fenrirscreenreader/commands/commands/subprocess.py

@@ -4,25 +4,42 @@
 # Fenrir TTY screen reader
 # By Chrys, Storm Dragon, and contributors.
 
-import _thread
 import os
 import subprocess
-from subprocess import PIPE
-from subprocess import Popen
 
 from fenrirscreenreader.core.i18n import _
 
 
+def run_script(script_path, current_user):
+    """Run an external command without accessing Fenrir's shared state."""
+    process = subprocess.Popen(
+        [script_path, current_user],
+        stdout=subprocess.PIPE,
+        stderr=subprocess.PIPE,
+        text=True,
+    )
+    stdout, stderr = process.communicate()
+    return {
+        "return_code": process.returncode,
+        "stdout": stdout,
+        "stderr": stderr,
+    }
+
+
 class command:
     def __init__(self):
-        pass
+        self._task_ids = set()
 
     def initialize(self, environment, script_path=""):
         self.env = environment
         self.script_path = script_path
 
     def shutdown(self):
-        pass
+        task_manager = self.env["runtime"].get("BackgroundTaskManager")
+        if task_manager:
+            for task_id in self._task_ids:
+                task_manager.cancel_task(task_id)
+        self._task_ids.clear()
 
     def get_description(self):
         return _("script: {0} fullpath: {1}").format(
@@ -49,29 +66,39 @@ class command:
                 interrupt=False,
             )
             return
-        _thread.start_new_thread(self._thread_run, ())
 
-    def _thread_run(self):
-        try:
-            callstring = (
-                self.script_path + " " + self.env["general"]["curr_user"]
-            )
-            p = Popen(callstring, stdout=PIPE, stderr=PIPE, shell=True)
-            stdout, stderr = p.communicate()
-            stdout = stdout.decode("utf-8")
-            stderr = stderr.decode("utf-8")
-            self.env["runtime"]["OutputManager"].interrupt_output()
-            if stderr != "":
-                self.env["runtime"]["OutputManager"].present_text(
-                    str(stderr), sound_icon="", interrupt=False
-                )
-            if stdout != "":
-                self.env["runtime"]["OutputManager"].present_text(
-                    str(stdout), sound_icon="", interrupt=False
-                )
-        except Exception as e:
+        task_id = self.env["runtime"][
+            "BackgroundTaskManager"
+        ].submit_external_task(
+            run_script,
+            self._handle_result,
+            self.script_path,
+            self.env["general"]["curr_user"],
+        )
+        if task_id is not None:
+            self._task_ids.add(task_id)
+
+    def _handle_result(self, result):
+        self._task_ids.discard(result.get("task_id"))
+        if not result.get("succeeded"):
             self.env["runtime"]["OutputManager"].present_text(
-                str(e), sound_icon="", interrupt=False
+                result.get("error", _("Script failed")),
+                sound_icon="",
+                interrupt=False,
+            )
+            return
+
+        output = result.get("value") or {}
+        stderr = output.get("stderr", "")
+        stdout = output.get("stdout", "")
+        self.env["runtime"]["OutputManager"].interrupt_output()
+        if stderr:
+            self.env["runtime"]["OutputManager"].present_text(
+                stderr, sound_icon="", interrupt=False
+            )
+        if stdout:
+            self.env["runtime"]["OutputManager"].present_text(
+                stdout, sound_icon="", interrupt=False
             )
 
     def set_callback(self, callback):

src/fenrirscreenreader/commands/commands/toggle_diff_mode.py

@@ -0,0 +1,27 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+
+# Fenrir TTY screen reader
+# By Chrys, Storm Dragon, and contributors.
+
+from fenrirscreenreader.core.i18n import _
+
+
+class command:
+    def __init__(self):
+        pass
+
+    def initialize(self, environment):
+        self.env = environment
+
+    def shutdown(self):
+        pass
+
+    def get_description(self):
+        return _("toggle diff review mode")
+
+    def run(self):
+        self.env["runtime"]["DiffReviewManager"].toggle_mode()
+
+    def set_callback(self, callback):
+        pass

src/fenrirscreenreader/commands/commands/toggle_tutorial_mode.py

@@ -19,19 +19,47 @@ class command:
         pass
 
     def get_description(self):
-        self.env["runtime"]["HelpManager"].toggle_tutorial_mode()
-        return _(
-            "Exiting tutorial mode. To enter tutorial mode again press Fenrir+f1"
-        )
+        return _("enter or leave tutorial mode")
 
     def run(self):
-        self.env["runtime"]["HelpManager"].toggle_tutorial_mode()
-        if self.env["runtime"]["HelpManager"].is_tutorial_mode():
+        help_manager = self.env["runtime"]["HelpManager"]
+        was_active = help_manager.is_tutorial_mode()
+        help_manager.toggle_tutorial_mode()
+        is_active = help_manager.is_tutorial_mode()
+        if was_active and is_active:
             self.env["runtime"]["OutputManager"].present_text(
                 _(
-                    "Entering tutorial mode. In this mode commands are described but not "
-                    "executed. You can move through the list of commands with the up and "
-                    "down arrow keys. To Exit tutorial mode press Fenrir+f1."
+                    "Unable to exit tutorial mode because exclusive keyboard "
+                    "capture could not be released. Press Fenrir+F1 or Escape "
+                    "to try again."
+                ),
+                interrupt=True,
+            )
+        elif is_active:
+            self.env["runtime"]["OutputManager"].present_text(
+                _(
+                    "Entering tutorial mode. In this mode commands are "
+                    "described but not executed. Use up and down to browse "
+                    "actions and left and right to switch between Active "
+                    "actions, Unbound actions, and Plugins. Press Space to "
+                    "repeat the current item. To exit tutorial mode press "
+                    "Fenrir+F1 or Escape. Active actions."
+                ),
+                interrupt=True,
+            )
+            if self.env["runtime"]["HelpManager"].is_capture_degraded():
+                self.env["runtime"]["OutputManager"].present_text(
+                    _(
+                        "Warning: full keyboard capture is unavailable. "
+                        "Unbound keys may reach the active application."
+                    ),
+                    interrupt=False,
+                )
+        else:
+            self.env["runtime"]["OutputManager"].present_text(
+                _(
+                    "Exiting tutorial mode. To enter tutorial mode again "
+                    "press Fenrir+F1"
                 ),
                 interrupt=True,
             )

src/fenrirscreenreader/commands/commands/toggle_vmenu_mode.py

@@ -19,7 +19,6 @@ class command:
         pass
 
     def get_description(self):
-        self.env["runtime"]["VmenuManager"].toggle_vmenu_mode()
         return _("Entering or Leaving v menu mode.")
 
     def run(self):

src/fenrirscreenreader/commands/commands/voice_browser.py

@@ -1,6 +1,5 @@
 #!/usr/bin/env python3
 
-import subprocess
 import time
 
 from fenrirscreenreader.core.i18n import _
@@ -8,11 +7,14 @@ from fenrirscreenreader.core.i18n import _
 
 class command:
     def __init__(self):
-        pass
+        self._request_generation = 0
+        self._loading = False
 
     def initialize(self, environment):
         self.env = environment
-        self.testMessage = "This is a voice test. The quick brown fox jumps over the lazy dog."
+        self.test_message = (
+            "This is a voice test. The quick brown fox jumps over the lazy dog."
+        )
         self.modules = []
         self.voices = []
         self.module_index = 0
@@ -22,7 +24,9 @@ class command:
         self.lastAnnounceTime = 0
 
     def shutdown(self):
-        pass
+        self._request_generation += 1
+        self._loading = False
+        self._leave_voice_browser(False)
 
     def get_description(self):
         return "Interactive voice browser with arrow key navigation"
@@ -36,34 +40,59 @@ class command:
             "Starting voice browser", interrupt=True
         )
 
-        # Load modules
-        self.modules = self.get_speechd_modules()
-        if not self.modules:
+        if self._loading:
+            return
+        self._loading = True
+        self._request_generation += 1
+        generation = self._request_generation
+        self.env["runtime"]["SpeechDiscoveryManager"].request_modules(
+            lambda modules, error: self._modules_ready(
+                generation, modules, error
+            )
+        )
+
+    def _modules_ready(self, generation, modules, error):
+        if generation != self._request_generation:
+            return
+        if error or not modules:
+            self._loading = False
             self.env["runtime"]["OutputManager"].present_text(
                 "No speech modules found", interrupt=True
             )
-            self.env["runtime"]["OutputManager"].play_sound("Error")
             return
-
-        # Set current module
+        self.modules = modules
         current_module = self.env["runtime"]["SettingsManager"].get_setting(
             "speech", "module"
         )
         if current_module and current_module in self.modules:
             self.module_index = self.modules.index(current_module)
+        self._request_current_module_voices(generation, True)
 
-        # Load voices
-        self.load_voices_for_current_module()
+    def _request_current_module_voices(self, generation, enter_browser=False):
+        module = self.modules[self.module_index]
+        self.env["runtime"]["OutputManager"].present_text(
+            f"Loading voices for {module}", interrupt=True
+        )
+        self.env["runtime"]["SpeechDiscoveryManager"].request_voices(
+            module,
+            lambda _module, voices, error: self._voices_ready(
+                generation, voices, error, enter_browser
+            ),
+        )
 
-        # Set current voice
+    def _voices_ready(self, generation, voices, error, enter_browser):
+        if generation != self._request_generation:
+            return
+        self._loading = False
+        self.voices = voices if not error else []
+        self.voice_index = 0
         current_voice = self.env["runtime"]["SettingsManager"].get_setting(
             "speech", "voice"
         )
         if current_voice and current_voice in self.voices:
             self.voice_index = self.voices.index(current_voice)
-
-        # Enter browser mode
-        self.enter_voice_browser()
+        if enter_browser:
+            self.enter_voice_browser()
         self.announce_current_selection()
 
     def enter_voice_browser(self):
@@ -112,6 +141,9 @@ class command:
 
     def exit_voice_browser(self):
         """Exit voice browser and restore normal key bindings"""
+        self._leave_voice_browser(True)
+
+    def _leave_voice_browser(self, announce):
         if not self.browserActive:
             return
 
@@ -124,16 +156,10 @@ class command:
         if "voiceBrowserInstance" in self.env["runtime"]:
             del self.env["runtime"]["voiceBrowserInstance"]
 
-        self.env["runtime"]["OutputManager"].present_text(
-            "Voice browser exited", interrupt=True
-        )
-
-    def load_voices_for_current_module(self):
-        """Load voices for current module"""
-        if self.module_index < len(self.modules):
-            module = self.modules[self.module_index]
-            self.voices = self.get_module_voices(module)
-            self.voice_index = 0  # Reset to first voice when changing modules
+        if announce:
+            self.env["runtime"]["OutputManager"].present_text(
+                "Voice browser exited", interrupt=True
+            )
 
     def announce_current_selection(self):
         """Announce current module and voice"""
@@ -148,7 +174,7 @@ class command:
 
         module = self.modules[self.module_index]
         if self.voices and self.voice_index < len(self.voices):
-            voice = self.voices[self.voice_index]
+            voice = self.voices[self.voice_index].split("|", 1)[0]
             self.env["runtime"]["OutputManager"].present_text(
                 f"{module}: {voice} ({self.voice_index + 1}/{len(self.voices)})",
                 interrupt=True,
@@ -173,16 +199,19 @@ class command:
         self.announce_current_selection()
 
     def next_module(self):
-        """Move to next module"""
-        self.module_index = (self.module_index + 1) % len(self.modules)
-        self.load_voices_for_current_module()
-        self.announce_current_selection()
+        self._change_module(1)
 
     def prev_module(self):
-        """Move to previous module"""
-        self.module_index = (self.module_index - 1) % len(self.modules)
-        self.load_voices_for_current_module()
-        self.announce_current_selection()
+        self._change_module(-1)
+
+    def _change_module(self, offset):
+        if self._loading or not self.modules:
+            return
+        self._loading = True
+        self._request_generation += 1
+        generation = self._request_generation
+        self.module_index = (self.module_index + offset) % len(self.modules)
+        self._request_current_module_voices(generation)
 
     def test_voice(self):
         """Test current voice"""
@@ -193,15 +222,38 @@ class command:
             return
 
         module = self.modules[self.module_index]
-        voice = self.voices[self.voice_index]
+        encoded_voice = self.voices[self.voice_index]
+        voice_name = encoded_voice.split("|", 1)[0]
 
         self.env["runtime"]["OutputManager"].present_text(
             "Testing...", interrupt=True
         )
-        if self.preview_voice(module, voice):
-            # Store for apply command
+        self._request_generation += 1
+        generation = self._request_generation
+        self._loading = True
+        self.env["runtime"]["SpeechDiscoveryManager"].request_voice_test(
+            module,
+            voice_name,
+            self.test_message,
+            lambda succeeded, error: self._voice_test_ready(
+                generation, module, encoded_voice, succeeded, error
+            ),
+        )
+
+    def _voice_test_ready(
+        self, generation, module, voice, succeeded, _error
+    ):
+        if generation != self._request_generation:
+            return
+        self._loading = False
+        if succeeded:
             self.env["commandBuffer"]["lastTestedModule"] = module
-            self.env["commandBuffer"]["lastTestedVoice"] = voice
+            voice_name, separator, language = voice.partition("|")
+            self.env["commandBuffer"]["lastTestedVoice"] = voice_name
+            if separator:
+                self.env["commandBuffer"]["lastTestedLanguage"] = language
+            else:
+                self.env["commandBuffer"].pop("lastTestedLanguage", None)
             self.env["runtime"]["OutputManager"].play_sound("Accept")
         else:
             self.env["runtime"]["OutputManager"].play_sound("Error")
@@ -212,13 +264,17 @@ class command:
             return
 
         module = self.modules[self.module_index]
-        voice = self.voices[self.voice_index]
+        voice, separator, language = self.voices[self.voice_index].partition(
+            "|"
+        )
 
         try:
             SettingsManager = self.env["runtime"]["SettingsManager"]
             SettingsManager.settings["speech"]["driver"] = "speechdDriver"
             SettingsManager.settings["speech"]["module"] = module
             SettingsManager.settings["speech"]["voice"] = voice
+            if separator:
+                SettingsManager.settings["speech"]["language"] = language
 
             if "SpeechDriver" in self.env["runtime"]:
                 SpeechDriver = self.env["runtime"]["SpeechDriver"]
@@ -236,66 +292,5 @@ class command:
             )
             self.env["runtime"]["OutputManager"].play_sound("Error")
 
-    def preview_voice(self, module, voice):
-        """Test voice with spd-say"""
-        try:
-            cmd = ["spd-say", "-o", module, "-y", voice, self.testMessage]
-            result = subprocess.run(cmd, timeout=10)
-            return result.returncode == 0
-        except Exception:
-            return False
-
-    def get_speechd_modules(self):
-        """Get available speech modules"""
-        try:
-            result = subprocess.run(
-                ["spd-say", "-O"], capture_output=True, text=True, timeout=10
-            )
-            if result.returncode == 0:
-                lines = result.stdout.strip().split("\n")
-                return [line.strip() for line in lines[1:] if line.strip()]
-        except Exception:
-            pass
-        return []
-
-    def get_module_voices(self, module):
-        """Get voices for module"""
-        try:
-            result = subprocess.run(
-                ["spd-say", "-o", module, "-L"],
-                capture_output=True,
-                text=True,
-                timeout=10,
-            )
-            if result.returncode == 0:
-                lines = result.stdout.strip().split("\n")
-                voices = []
-                for line in lines[1:]:
-                    if not line.strip():
-                        continue
-                    if module.lower() == "espeak-ng":
-                        voice = self.process_espeak_voice(line)
-                        if voice:
-                            voices.append(voice)
-                    else:
-                        voices.append(line.strip())
-                return voices
-        except Exception:
-            pass
-        return []
-
-    def process_espeak_voice(self, voiceLine):
-        """Process espeak voice format"""
-        parts = [p for p in voiceLine.split() if p]
-        if len(parts) < 2:
-            return None
-        lang_code = parts[-2].lower()
-        variant = parts[-1].lower()
-        return (
-            f"{lang_code}+{variant}"
-            if variant and variant != "none"
-            else lang_code
-        )
-
     def set_callback(self, callback):
         pass

src/fenrirscreenreader/commands/commands/voice_browser_safe.py

@@ -1,225 +1,160 @@
 #!/usr/bin/env python3
 
-import subprocess
-import threading
-import time
-
+from fenrirscreenreader.core import debug
 from fenrirscreenreader.core.i18n import _
 
 
 class command:
     def __init__(self):
-        pass
+        self._request_generation = 0
+        self._loading = False
 
     def initialize(self, environment):
         self.env = environment
-        self.testMessage = (
+        self.test_message = (
             "Voice test: The quick brown fox jumps over the lazy dog."
         )
 
     def shutdown(self):
-        pass
+        self._request_generation += 1
+        self._loading = False
 
     def get_description(self):
-        return "Safe voice browser - cycles through voices without hanging"
+        return _("browse and test Speech Dispatcher voices")
 
     def run(self):
-        try:
+        if self._loading:
+            return
+        self._loading = True
+        self._request_generation += 1
+        generation = self._request_generation
+        self.env["runtime"]["OutputManager"].present_text(
+            _("Loading speech modules"), interrupt=True
+        )
+        self.env["runtime"]["SpeechDiscoveryManager"].request_modules(
+            lambda modules, error: self._modules_ready(
+                generation, modules, error
+            )
+        )
+
+    def _modules_ready(self, generation, modules, error):
+        if generation != self._request_generation:
+            return
+        if error or not modules:
+            self._loading = False
+            self._report_error(_("No speech modules found"), error)
+            return
+
+        module_index = self.env["commandBuffer"].get(
+            "safeBrowserModuleIndex", 0
+        )
+        if module_index >= len(modules):
+            module_index = 0
+        module = modules[module_index]
+        self.env["runtime"]["OutputManager"].present_text(
+            _("Loading voices for {module}").format(module=module),
+            interrupt=True,
+        )
+        self.env["runtime"]["SpeechDiscoveryManager"].request_voices(
+            module,
+            lambda _module, voices, voice_error: self._voices_ready(
+                generation,
+                modules,
+                module_index,
+                voices,
+                voice_error,
+            ),
+        )
+
+    def _voices_ready(
+        self, generation, modules, module_index, voices, error
+    ):
+        if generation != self._request_generation:
+            return
+        if error or not voices:
+            self._loading = False
+            next_module = (module_index + 1) % len(modules)
+            self.env["commandBuffer"]["safeBrowserModuleIndex"] = next_module
+            self.env["commandBuffer"]["safeBrowserVoiceIndex"] = 0
+            self._report_error(_("No voices found"), error)
+            return
+
+        voice_index = self.env["commandBuffer"].get(
+            "safeBrowserVoiceIndex", 0
+        )
+        if voice_index >= len(voices):
+            voice_index = 0
+        module = modules[module_index]
+        voice = voices[voice_index]
+        voice_name = voice.split("|", 1)[0]
+        self.env["runtime"]["OutputManager"].present_text(
+            _("Module: {module}, voice: {voice}").format(
+                module=module, voice=voice_name
+            ),
+            interrupt=True,
+        )
+        self.env["runtime"]["OutputManager"].present_text(
+            _("Testing voice"), interrupt=True
+        )
+        self.env["runtime"]["SpeechDiscoveryManager"].request_voice_test(
+            module,
+            voice_name,
+            self.test_message,
+            lambda succeeded, test_error: self._test_ready(
+                generation,
+                modules,
+                module_index,
+                voices,
+                voice_index,
+                succeeded,
+                test_error,
+            ),
+        )
+
+    def _test_ready(
+        self,
+        generation,
+        modules,
+        module_index,
+        voices,
+        voice_index,
+        succeeded,
+        error,
+    ):
+        if generation != self._request_generation:
+            return
+        self._loading = False
+        module = modules[module_index]
+        voice_name, separator, language = voices[voice_index].partition("|")
+        if succeeded:
+            self.env["commandBuffer"]["lastTestedModule"] = module
+            self.env["commandBuffer"]["lastTestedVoice"] = voice_name
+            if separator:
+                self.env["commandBuffer"]["lastTestedLanguage"] = language
+            else:
+                self.env["commandBuffer"].pop("lastTestedLanguage", None)
             self.env["runtime"]["OutputManager"].present_text(
-                "Starting safe voice browser", interrupt=True
+                _("Voice test completed successfully"), interrupt=True
             )
+        else:
+            self._report_error(_("Voice test failed"), error)
 
-            # Get modules with timeout protection
-            modules = self.get_speechd_modules_with_timeout()
-            if not modules:
-                self.env["runtime"]["OutputManager"].present_text(
-                    "No speech modules found", interrupt=True
-                )
-                return
+        voice_index += 1
+        if voice_index >= len(voices):
+            voice_index = 0
+            module_index = (module_index + 1) % len(modules)
+        self.env["commandBuffer"]["safeBrowserModuleIndex"] = module_index
+        self.env["commandBuffer"]["safeBrowserVoiceIndex"] = voice_index
 
-            # Get current position from commandBuffer or start fresh
-            module_index = self.env["commandBuffer"].get(
-                "safeBrowserModuleIndex", 0
+    def _report_error(self, message, detail):
+        if detail:
+            self.env["runtime"]["DebugManager"].write_debug_out(
+                f"voice_browser_safe: {detail}",
+                debug.DebugLevel.ERROR,
             )
-            voice_index = self.env["commandBuffer"].get(
-                "safeBrowserVoiceIndex", 0
-            )
-
-            # Ensure valid module index
-            if module_index >= len(modules):
-                module_index = 0
-
-            current_module = modules[module_index]
-            self.env["runtime"]["OutputManager"].present_text(
-                f"Loading voices for {current_module}...", interrupt=True
-            )
-
-            # Get voices with timeout protection
-            voices = self.get_module_voices_with_timeout(current_module)
-            if not voices:
-                self.env["runtime"]["OutputManager"].present_text(
-                    f"No voices in {current_module}, trying next module",
-                    interrupt=True,
-                )
-                module_index = (module_index + 1) % len(modules)
-                self.env["commandBuffer"][
-                    "safeBrowserModuleIndex"
-                ] = module_index
-                self.env["commandBuffer"]["safeBrowserVoiceIndex"] = 0
-                return
-
-            # Ensure valid voice index
-            if voice_index >= len(voices):
-                voice_index = 0
-
-            current_voice = voices[voice_index]
-
-            # Announce current selection
-            self.env["runtime"]["OutputManager"].present_text(
-                f"Module: {current_module} ({module_index + 1}/{len(modules)})",
-                interrupt=True,
-            )
-            self.env["runtime"]["OutputManager"].present_text(
-                f"Voice: {current_voice} ({voice_index + 1}/{len(voices)})",
-                interrupt=True,
-            )
-
-            # Test voice in background thread to avoid blocking
-            self.env["runtime"]["OutputManager"].present_text(
-                "Testing voice...", interrupt=True
-            )
-
-            # Use threading to prevent freezing
-            test_thread = threading.Thread(
-                target=self.test_voice_async,
-                args=(current_module, current_voice),
-            )
-            test_thread.daemon = True
-            test_thread.start()
-
-            # Store tested voice for apply command
-            self.env["commandBuffer"]["lastTestedModule"] = current_module
-            self.env["commandBuffer"]["lastTestedVoice"] = current_voice
-
-            # Advance to next voice for next run
-            voice_index += 1
-            if voice_index >= len(voices):
-                voice_index = 0
-                module_index = (module_index + 1) % len(modules)
-
-            # Store position for next run
-            self.env["commandBuffer"]["safeBrowserModuleIndex"] = module_index
-            self.env["commandBuffer"]["safeBrowserVoiceIndex"] = voice_index
-
-            # Give instructions
-            self.env["runtime"]["OutputManager"].present_text(
-                "Run again for next voice, or use apply voice command",
-                interrupt=True,
-            )
-
-        except Exception as e:
-            self.env["runtime"]["OutputManager"].present_text(
-                f"Voice browser error: {str(e)}", interrupt=True
-            )
-            self.env["runtime"]["OutputManager"].play_sound("Error")
-
-    def test_voice_async(self, module, voice):
-        """Test voice in background thread to avoid blocking"""
-        try:
-            # Run with strict timeout
-            cmd = ["spd-say", "-o", module, "-y", voice, self.testMessage]
-            result = subprocess.run(cmd, timeout=5, capture_output=True)
-
-            # Schedule success sound for main thread
-            if result.returncode == 0:
-                # We can't call OutputManager from background thread safely
-                # So we'll just let the main thread handle feedback
-                pass
-
-        except subprocess.TimeoutExpired:
-            # Voice test timed out - this is okay, don't crash
-            pass
-        except Exception:
-            # Any other error - also okay, don't crash
-            pass
-
-    def get_speechd_modules_with_timeout(self):
-        """Get speech modules with timeout protection"""
-        try:
-            result = subprocess.run(
-                ["spd-say", "-O"], capture_output=True, text=True, timeout=3
-            )
-            if result.returncode == 0:
-                lines = result.stdout.strip().split("\n")
-                modules = [line.strip() for line in lines[1:] if line.strip()]
-                # Limit to first 10 modules to prevent overload
-                return modules[:10]
-        except subprocess.TimeoutExpired:
-            self.env["runtime"]["OutputManager"].present_text(
-                "Module detection timed out", interrupt=True
-            )
-        except Exception as e:
-            self.env["runtime"]["OutputManager"].present_text(
-                f"Module detection failed: {str(e)}", interrupt=True
-            )
-        return []
-
-    def get_module_voices_with_timeout(self, module):
-        """Get voices with timeout and limits"""
-        try:
-            result = subprocess.run(
-                ["spd-say", "-o", module, "-L"],
-                capture_output=True,
-                text=True,
-                timeout=5,
-            )
-            if result.returncode == 0:
-                lines = result.stdout.strip().split("\n")
-                voices = []
-                for line in lines[1:]:
-                    if not line.strip():
-                        continue
-                    if module.lower() == "espeak-ng":
-                        voice = self.process_espeak_voice(line)
-                        if voice:
-                            voices.append(voice)
-                    else:
-                        voices.append(line.strip())
-
-                # Limit voice count to prevent memory issues
-                if len(voices) > 1000:
-                    self.env["runtime"]["OutputManager"].present_text(
-                        f"found {len(voices)} voices, limiting to first 1000",
-                        interrupt=True,
-                    )
-                    voices = voices[:1000]
-
-                return voices
-        except subprocess.TimeoutExpired:
-            self.env["runtime"]["OutputManager"].present_text(
-                f"Voice detection for {module} timed out", interrupt=True
-            )
-        except Exception as e:
-            self.env["runtime"]["OutputManager"].present_text(
-                f"Voice detection failed: {str(e)}", interrupt=True
-            )
-        return []
-
-    def process_espeak_voice(self, voiceLine):
-        """Process espeak voice format"""
-        try:
-            parts = [p for p in voiceLine.split() if p]
-            if len(parts) < 2:
-                return None
-            lang_code = parts[-2].lower()
-            variant = parts[-1].lower()
-            return (
-                f"{lang_code}+{variant}"
-                if variant and variant != "none"
-                else lang_code
-            )
-        except Exception:
-            return None
+        self.env["runtime"]["OutputManager"].present_text(
+            message, interrupt=True
+        )
 
     def set_callback(self, callback):
         pass

src/fenrirscreenreader/commands/help/next_help_section.py

@@ -0,0 +1,30 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+
+# Fenrir TTY screen reader
+# By Chrys, Storm Dragon, and contributors.
+
+from fenrirscreenreader.core.i18n import _
+
+
+class command:
+    def __init__(self):
+        pass
+
+    def initialize(self, environment):
+        self.env = environment
+
+    def shutdown(self):
+        pass
+
+    def get_description(self):
+        return _("show next help category")
+
+    def run(self):
+        text = self.env["runtime"]["HelpManager"].next_help_section()
+        self.env["runtime"]["OutputManager"].present_text(
+            text, interrupt=True
+        )
+
+    def set_callback(self, callback):
+        pass

src/fenrirscreenreader/commands/help/prev_help_section.py

@@ -0,0 +1,30 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+
+# Fenrir TTY screen reader
+# By Chrys, Storm Dragon, and contributors.
+
+from fenrirscreenreader.core.i18n import _
+
+
+class command:
+    def __init__(self):
+        pass
+
+    def initialize(self, environment):
+        self.env = environment
+
+    def shutdown(self):
+        pass
+
+    def get_description(self):
+        return _("show previous help category")
+
+    def run(self):
+        text = self.env["runtime"]["HelpManager"].prev_help_section()
+        self.env["runtime"]["OutputManager"].present_text(
+            text, interrupt=True
+        )
+
+    def set_callback(self, callback):
+        pass

src/fenrirscreenreader/commands/onByteInput/10000-shut_up.py

@@ -1,55 +0,0 @@
-#!/usr/bin/env python3
-# -*- coding: utf-8 -*-
-
-# Fenrir TTY screen reader
-# By Chrys, Storm Dragon, and contributors.
-
-
-from fenrirscreenreader.core.i18n import _
-
-
-class command:
-    def __init__(self):
-        pass
-
-    def initialize(self, environment):
-        self.env = environment
-
-    def shutdown(self):
-        pass
-
-    def get_description(self):
-        return ""
-
-    def run(self):
-        if not self.env["runtime"]["SettingsManager"].get_setting_as_bool(
-            "keyboard", "interrupt_on_key_press"
-        ):
-            return
-        if self.env["runtime"]["InputManager"].no_key_pressed():
-            return
-        if self.env["runtime"]["ScreenManager"].is_screen_change():
-            return
-        if len(self.env["input"]["curr_input"]) <= len(
-            self.env["input"]["prev_input"]
-        ):
-            return
-        # if the filter is set
-        if (
-            self.env["runtime"]["SettingsManager"]
-            .get_setting("keyboard", "interrupt_on_key_press_filter")
-            .strip()
-            != ""
-        ):
-            filter_list = (
-                self.env["runtime"]["SettingsManager"]
-                .get_setting("keyboard", "interrupt_on_key_press_filter")
-                .split(",")
-            )
-            for curr_key in self.env["input"]["curr_input"]:
-                if curr_key not in filter_list:
-                    return
-        self.env["runtime"]["OutputManager"].interrupt_output()
-
-    def set_callback(self, callback):
-        pass

src/fenrirscreenreader/commands/onByteInput/15000-enable_temp_speech.py

@@ -1,41 +0,0 @@
-#!/usr/bin/env python3
-# -*- coding: utf-8 -*-
-
-# Fenrir TTY screen reader
-# By Chrys, Storm Dragon, and contributors.
-
-
-from fenrirscreenreader.core.i18n import _
-
-
-class command:
-    def __init__(self):
-        pass
-
-    def initialize(self, environment):
-        self.env = environment
-
-    def shutdown(self):
-        pass
-
-    def get_description(self):
-        return _("disables speech until next keypress")
-
-    def run(self):
-        if not self.env["commandBuffer"]["enableSpeechOnKeypress"]:
-            return
-        self.env["runtime"]["SettingsManager"].set_setting(
-            "speech",
-            "enabled",
-            str(self.env["commandBuffer"]["enableSpeechOnKeypress"]),
-        )
-        self.env["commandBuffer"]["enableSpeechOnKeypress"] = False
-        # Also disable prompt watching since speech was manually re-enabled
-        if "silenceUntilPrompt" in self.env["commandBuffer"]:
-            self.env["commandBuffer"]["silenceUntilPrompt"] = False
-        self.env["runtime"]["OutputManager"].present_text(
-            _("speech enabled"), sound_icon="SpeechOn", interrupt=True
-        )
-
-    def set_callback(self, callback):
-        pass

src/fenrirscreenreader/commands/onCursorChange/15000-char_echo.py

@@ -42,19 +42,10 @@ class command:
         )
         if x_move > 3:
             return
-        if self.env["runtime"]["InputManager"].get_shortcut_type() in ["KEY"]:
-            if self.env["runtime"][
-                "InputManager"
-            ].get_last_deepest_input() in [["KEY_TAB"]]:
-                return
-        elif self.env["runtime"]["InputManager"].get_shortcut_type() in [
-            "BYTE"
-        ]:
-            if self.env["runtime"]["ByteManager"].get_last_byte_key() in [
-                b"	",
-                b"\t",
-            ]:
-                return
+        if self.env["runtime"][
+            "InputManager"
+        ].get_last_deepest_input() in [["KEY_TAB"]]:
+            return
         # detect deletion or chilling
         if (
             self.env["screen"]["new_cursor"]["x"]

src/fenrirscreenreader/commands/onCursorChange/50000-present_char_if_cursor_change_horizontal.py

@@ -81,11 +81,10 @@ class command:
         if curr_char.isspace():
             # Only announce spaces during pure navigation (arrow keys)
             # Check if this is really navigation by looking at input history
-            if self.env["runtime"]["InputManager"].get_shortcut_type() in [
-                "KEY"
-            ] and self.env["runtime"]["InputManager"].get_last_deepest_input()[
-                0
-            ] in [
+            last_input = self.env["runtime"][
+                "InputManager"
+            ].get_last_deepest_input()
+            if last_input and last_input[0] in [
                 "KEY_LEFT",
                 "KEY_RIGHT",
                 "KEY_UP",

src/fenrirscreenreader/commands/onCursorChange/54000-tab_completion.py

@@ -0,0 +1,44 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+
+# Fenrir TTY screen reader
+# By Chrys, Storm Dragon, and contributors.
+
+from fenrirscreenreader.core.i18n import _
+from fenrirscreenreader.core.tabCompletionManager import TabCompletionManager
+
+
+class command:
+    def __init__(self):
+        self.manager = TabCompletionManager()
+
+    def initialize(self, environment):
+        self.env = environment
+        self.manager.initialize(environment)
+
+    def shutdown(self):
+        pass
+
+    def get_description(self):
+        return _("Announces tab completions on cursor movement")
+
+    def run(self):
+        text = self.manager.process_update()
+        if not text:
+            return
+
+        do_interrupt = True
+        if self.env["runtime"]["SettingsManager"].get_setting_as_bool(
+            "speech", "auto_read_incoming"
+        ):
+            do_interrupt = False
+
+        self.env["runtime"]["OutputManager"].present_text(
+            text,
+            interrupt=do_interrupt,
+            announce_capital=True,
+            flush=False,
+        )
+
+    def set_callback(self, callback):
+        pass

src/fenrirscreenreader/commands/onCursorChange/55000-large_insertion_echo.py

@@ -0,0 +1,95 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+
+# Fenrir TTY screen reader
+# By Chrys, Storm Dragon, and contributors.
+
+import time
+
+from fenrirscreenreader.core.i18n import _
+
+
+class command:
+    def __init__(self):
+        pass
+
+    def initialize(self, environment):
+        self.env = environment
+
+    def shutdown(self):
+        pass
+
+    def get_description(self):
+        return _("Announces large text insertions at the cursor")
+
+    def run(self):
+        if self.env["runtime"]["ScreenManager"].is_screen_change():
+            return
+        if not self.env["runtime"]["ScreenManager"].is_delta():
+            return
+        if not self.env["screen"].get("new_delta_is_typing", False):
+            return
+
+        x_move = (
+            self.env["screen"]["new_cursor"]["x"]
+            - self.env["screen"]["old_cursor"]["x"]
+        )
+        if x_move < 5:
+            return
+        if self._is_recent_tab_input():
+            return
+
+        delta_text = self.env["screen"]["new_delta"]
+        if not self._matches_cursor_insert(x_move, delta_text):
+            return
+
+        curr_delta = delta_text
+        if (
+            len(curr_delta.strip()) != len(curr_delta)
+            and curr_delta.strip() != ""
+        ):
+            curr_delta = curr_delta.strip()
+        if not curr_delta:
+            return
+
+        do_interrupt = True
+        if self.env["runtime"]["SettingsManager"].get_setting_as_bool(
+            "speech", "auto_read_incoming"
+        ):
+            do_interrupt = False
+
+        self.env["runtime"]["OutputManager"].present_text(
+            curr_delta,
+            interrupt=do_interrupt,
+            announce_capital=True,
+            flush=False,
+        )
+
+    def _is_recent_tab_input(self):
+        input_manager = self.env["runtime"].get("InputManager")
+        if not input_manager:
+            return False
+        if input_manager.get_last_deepest_input() in [["KEY_TAB"]]:
+            return True
+
+        last_event = input_manager.get_last_event()
+        if not last_event or last_event.get("event_name") != "KEY_TAB":
+            return False
+        try:
+            return time.time() - input_manager.get_last_input_time() <= 0.5
+        except Exception:
+            return False
+
+    def _matches_cursor_insert(self, x_move, delta_text):
+        if not delta_text or "\n" in delta_text:
+            return False
+
+        delta_len = len(delta_text)
+        if x_move == delta_len:
+            return True
+        if abs(x_move - delta_len) <= 2:
+            return True
+        return delta_len > 10 and abs(x_move - delta_len) <= (delta_len * 0.2)
+
+    def set_callback(self, callback):
+        pass

src/fenrirscreenreader/commands/onCursorChange/55000-tab_completion.py

@@ -1,170 +0,0 @@
-#!/usr/bin/env python3
-# -*- coding: utf-8 -*-
-
-# Fenrir TTY screen reader
-# By Chrys, Storm Dragon, and contributors.
-
-import time
-from fenrirscreenreader.core.i18n import _
-
-
-class command:
-    def __init__(self):
-        pass
-
-    def initialize(self, environment):
-        self.env = environment
-        # Initialize tab completion state tracking
-        if "tabCompletion" not in self.env["commandBuffer"]:
-            self.env["commandBuffer"]["tabCompletion"] = {
-                "lastTabTime": 0,
-                "pendingCompletion": None,
-                "retryCount": 0
-            }
-
-    def shutdown(self):
-        pass
-
-    def get_description(self):
-        return _("Announces tab completions when detected")
-
-    def _is_recent_tab_input(self):
-        """Check if TAB was pressed recently (within 200ms window)"""
-        current_time = time.time()
-        tab_detected = False
-        
-        # Check KEY mode
-        if self.env["runtime"]["InputManager"].get_shortcut_type() in ["KEY"]:
-            if (self.env["runtime"]["InputManager"].get_last_deepest_input()
-                in [["KEY_TAB"]]):
-                tab_detected = True
-                self.env["commandBuffer"]["tabCompletion"]["lastTabTime"] = current_time
-        
-        # Check BYTE mode
-        elif self.env["runtime"]["InputManager"].get_shortcut_type() in ["BYTE"]:
-            for currByte in self.env["runtime"]["ByteManager"].get_last_byte_key():
-                if currByte == 9:  # Tab character
-                    tab_detected = True
-                    self.env["commandBuffer"]["tabCompletion"]["lastTabTime"] = current_time
-        
-        # Check if tab was pressed recently (200ms window)
-        if not tab_detected:
-            time_since_tab = current_time - self.env["commandBuffer"]["tabCompletion"]["lastTabTime"]
-            if time_since_tab <= 0.2:  # 200ms window
-                tab_detected = True
-        
-        return tab_detected
-
-    def _is_flexible_completion_match(self, x_move, delta_text):
-        """Use flexible matching instead of strict equality"""
-        if not delta_text:
-            return False
-        
-        delta_len = len(delta_text)
-        
-        # Exact match (preserve original behavior)
-        if x_move == delta_len:
-            return True
-        
-        # Flexible range: allow ±2 characters difference
-        # Handles spacing adjustments and unicode width variations
-        if abs(x_move - delta_len) <= 2 and delta_len > 0:
-            return True
-        
-        # For longer completions, allow proportional variance
-        if delta_len > 10 and abs(x_move - delta_len) <= (delta_len * 0.2):
-            return True
-        
-        return False
-
-    def _detect_completion_patterns(self, delta_text):
-        """Detect common tab completion patterns for improved accuracy"""
-        if not delta_text:
-            return False
-        
-        delta_stripped = delta_text.strip()
-        
-        # File extension completion
-        if '.' in delta_stripped and delta_stripped.count('.') <= 2:
-            return True
-        
-        # Path completion (contains / or \)
-        if '/' in delta_stripped or '\\' in delta_stripped:
-            return True
-        
-        # Command parameter completion (starts with -)
-        if delta_stripped.startswith('-') and len(delta_stripped) > 1:
-            return True
-        
-        # Word boundary completion (alphanumeric content)
-        if delta_stripped.isalnum() and len(delta_stripped) >= 2:
-            return True
-        
-        return False
-
-    def run(self):
-        """Enhanced tab completion detection with improved reliability"""
-        # Basic cursor movement check (preserve original logic)
-        x_move = (
-            self.env["screen"]["new_cursor"]["x"]
-            - self.env["screen"]["old_cursor"]["x"]
-        )
-        if x_move <= 0:
-            return
-
-        # Enhanced tab input detection with persistence
-        tab_detected = self._is_recent_tab_input()
-        
-        # Fallback for non-tab movements (preserve original thresholds)
-        if not tab_detected:
-            if x_move < 5:
-                return
-
-        # Screen delta availability check
-        if not self.env["runtime"]["ScreenManager"].is_delta():
-            # If tab was detected but no delta yet, store for potential retry
-            if tab_detected and self.env["commandBuffer"]["tabCompletion"]["retryCount"] < 2:
-                self.env["commandBuffer"]["tabCompletion"]["pendingCompletion"] = {
-                    "x_move": x_move,
-                    "timestamp": time.time()
-                }
-                self.env["commandBuffer"]["tabCompletion"]["retryCount"] += 1
-            return
-
-        delta_text = self.env["screen"]["new_delta"]
-        
-        # Enhanced correlation checking with flexible matching
-        if not self._is_flexible_completion_match(x_move, delta_text):
-            # Additional pattern-based validation for edge cases
-            if not (tab_detected and self._detect_completion_patterns(delta_text)):
-                return
-
-        # Reset retry counter on successful detection
-        self.env["commandBuffer"]["tabCompletion"]["retryCount"] = 0
-        self.env["commandBuffer"]["tabCompletion"]["pendingCompletion"] = None
-
-        # Mark that we've handled this delta to prevent duplicate announcements
-        # This prevents the incoming text handler from also announcing the same content
-        self.env["commandBuffer"]["tabCompletion"]["lastProcessedDelta"] = delta_text
-        self.env["commandBuffer"]["tabCompletion"]["lastProcessedTime"] = time.time()
-
-        # Text filtering and announcement (preserve original behavior)
-        curr_delta = delta_text
-        if (len(curr_delta.strip()) != len(curr_delta) and curr_delta.strip() != ""):
-            curr_delta = curr_delta.strip()
-
-        # Don't interrupt ongoing auto-read announcements
-        do_interrupt = True
-        if self.env["runtime"]["SettingsManager"].get_setting_as_bool(
-            "speech", "auto_read_incoming"
-        ):
-            do_interrupt = False
-
-        # Enhanced announcement with better handling of empty completions
-        if curr_delta:
-            self.env["runtime"]["OutputManager"].present_text(
-                curr_delta, interrupt=do_interrupt, announce_capital=True, flush=False
-            )
-
-    def set_callback(self, callback):
-        pass

src/fenrirscreenreader/commands/onHeartBeat/55000-tab_completion_retry.py

@@ -1,128 +0,0 @@
-#!/usr/bin/env python3
-# -*- coding: utf-8 -*-
-
-# Fenrir TTY screen reader
-# By Chrys, Storm Dragon, and contributors.
-
-import time
-from fenrirscreenreader.core.i18n import _
-
-
-class command:
-    def __init__(self):
-        pass
-
-    def initialize(self, environment):
-        self.env = environment
-
-    def shutdown(self):
-        pass
-
-    def get_description(self):
-        return _("Handles delayed retry for tab completion detection")
-
-    def run(self):
-        """Check for and process pending tab completions with slight delay"""
-        # Only process if we have tab completion state
-        if "tabCompletion" not in self.env["commandBuffer"]:
-            return
-
-        tab_state = self.env["commandBuffer"]["tabCompletion"]
-        pending = tab_state.get("pendingCompletion")
-        
-        if not pending:
-            return
-
-        current_time = time.time()
-        
-        # Process pending completion after 50ms delay
-        if current_time - pending["timestamp"] < 0.05:
-            return
-
-        # Check if screen delta is now available
-        if not self.env["runtime"]["ScreenManager"].is_delta():
-            # Give up after 200ms total
-            if current_time - pending["timestamp"] > 0.2:
-                tab_state["pendingCompletion"] = None
-                tab_state["retryCount"] = 0
-            return
-
-        # Process the delayed completion
-        delta_text = self.env["screen"]["new_delta"]
-        x_move = pending["x_move"]
-        
-        # Use the same flexible matching logic as main tab completion
-        match_found = self._is_flexible_completion_match(x_move, delta_text)
-        
-        if not match_found:
-            # Try pattern-based detection as final fallback
-            match_found = self._detect_completion_patterns(delta_text)
-
-        if match_found and delta_text:
-            # Mark that we've handled this delta to prevent duplicate announcements
-            tab_state["lastProcessedDelta"] = delta_text
-            tab_state["lastProcessedTime"] = current_time
-            
-            # Filter and announce the completion
-            curr_delta = delta_text
-            if (len(curr_delta.strip()) != len(curr_delta) and 
-                curr_delta.strip() != ""):
-                curr_delta = curr_delta.strip()
-            
-            if curr_delta:
-                self.env["runtime"]["OutputManager"].present_text(
-                    curr_delta, interrupt=True, announce_capital=True, flush=False
-                )
-
-        # Clear pending completion
-        tab_state["pendingCompletion"] = None
-        tab_state["retryCount"] = 0
-
-    def _is_flexible_completion_match(self, x_move, delta_text):
-        """Use flexible matching (duplicated from main command for heartbeat use)"""
-        if not delta_text:
-            return False
-        
-        delta_len = len(delta_text)
-        
-        # Exact match
-        if x_move == delta_len:
-            return True
-        
-        # Flexible range: allow ±2 characters difference
-        if abs(x_move - delta_len) <= 2 and delta_len > 0:
-            return True
-        
-        # For longer completions, allow proportional variance
-        if delta_len > 10 and abs(x_move - delta_len) <= (delta_len * 0.2):
-            return True
-        
-        return False
-
-    def _detect_completion_patterns(self, delta_text):
-        """Detect common tab completion patterns (duplicated from main command)"""
-        if not delta_text:
-            return False
-        
-        delta_stripped = delta_text.strip()
-        
-        # File extension completion
-        if '.' in delta_stripped and delta_stripped.count('.') <= 2:
-            return True
-        
-        # Path completion
-        if '/' in delta_stripped or '\\' in delta_stripped:
-            return True
-        
-        # Command parameter completion
-        if delta_stripped.startswith('-') and len(delta_stripped) > 1:
-            return True
-        
-        # Word boundary completion
-        if delta_stripped.isalnum() and len(delta_stripped) >= 2:
-            return True
-        
-        return False
-
-    def set_callback(self, callback):
-        pass

src/fenrirscreenreader/commands/onHeartBeat/76000-time.py

@@ -5,11 +5,17 @@
 # By Chrys, Storm Dragon, and contributors.
 
 import datetime
+import fcntl
+import os
+import tempfile
 import time
 
 from fenrirscreenreader.core.i18n import _
 
 
+ANNOUNCEMENT_LOCK_TIMEOUT_SEC = 5.0
+
+
 class command:
     def __init__(self):
         pass
@@ -26,6 +32,49 @@ class command:
     def get_description(self):
         return "No Description found"
 
+    def _get_announcement_lock_path(self):
+        return os.path.join(
+            tempfile.gettempdir(),
+            "fenrirscreenreader-time-announcement.lock",
+        )
+
+    def _read_announcement_lock_slot(self, lock_file):
+        lock_file.seek(0)
+        lock_content = lock_file.readline().strip().split()
+        if len(lock_content) < 2:
+            return ""
+        return lock_content[1]
+
+    def _claim_announcement_lock(self, announcement_slot):
+        now = time.time()
+        lock_path = self._get_announcement_lock_path()
+        try:
+            lock_fd = os.open(lock_path, os.O_CREAT | os.O_RDWR, 0o666)
+        except OSError:
+            return False
+        try:
+            os.chmod(lock_path, 0o666)
+        except OSError:
+            pass
+
+        with os.fdopen(lock_fd, "r+", encoding="utf-8") as lock_file:
+            try:
+                fcntl.flock(lock_file.fileno(), fcntl.LOCK_EX)
+                lock_slot = self._read_announcement_lock_slot(lock_file)
+                lock_stat = os.fstat(lock_file.fileno())
+
+                if lock_slot == announcement_slot:
+                    return False
+
+                lock_file.seek(0)
+                lock_file.truncate()
+                lock_file.write(f"{os.getpid()} {announcement_slot} {now}\n")
+                lock_file.flush()
+                os.fsync(lock_file.fileno())
+                return True
+            finally:
+                fcntl.flock(lock_file.fileno(), fcntl.LOCK_UN)
+
     def run(self):
         if not self.env["runtime"]["SettingsManager"].get_setting_as_bool(
             "time", "enabled"
@@ -50,6 +99,7 @@ class command:
         if delay_sec > 0:
             if int((now - self.last_time).total_seconds()) < delay_sec:
                 return
+            announcement_slot = f"delay:{int(now.timestamp()) // delay_sec}"
         else:
             # should announce?
             if not str(now.minute).zfill(2) in on_minutes:
@@ -58,6 +108,7 @@ class command:
             if now.hour == self.last_time.hour:
                 if now.minute == self.last_time.minute:
                     return
+            announcement_slot = f"minute:{datetime.datetime.strftime(now, '%Y%m%d%H%M')}"
 
         date_format = self.env["runtime"]["SettingsManager"].get_setting(
             "general", "date_format"
@@ -78,6 +129,10 @@ class command:
         if not (present_date or present_time):
             return
 
+        if not self._claim_announcement_lock(announcement_slot):
+            self.last_time = now
+            return
+
         time_format = self.env["runtime"]["SettingsManager"].get_setting(
             "general", "time_format"
         )

src/fenrirscreenreader/commands/onKeyInput/10000-shut_up.py

@@ -48,7 +48,7 @@ class command:
             for curr_key in self.env["input"]["curr_input"]:
                 if curr_key not in filter_list:
                     return
-        self.env["runtime"]["OutputManager"].interrupt_output()
+        self.env["runtime"]["OutputManager"].interrupt_output_async()
 
     def set_callback(self, callback):
         pass

src/fenrirscreenreader/commands/onKeyInput/55000-tab_completion_capture.py

@@ -0,0 +1,29 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+
+# Fenrir TTY screen reader
+# By Chrys, Storm Dragon, and contributors.
+
+from fenrirscreenreader.core.i18n import _
+from fenrirscreenreader.core.tabCompletionManager import TabCompletionManager
+
+
+class command:
+    def __init__(self):
+        self.manager = TabCompletionManager()
+
+    def initialize(self, environment):
+        self.env = environment
+        self.manager.initialize(environment)
+
+    def shutdown(self):
+        pass
+
+    def get_description(self):
+        return _("Tracks tab keypresses for completion announcements")
+
+    def run(self):
+        self.manager.capture_if_tab()
+
+    def set_callback(self, callback):
+        pass

src/fenrirscreenreader/commands/onKeyInput/80000-capslock.py

@@ -27,6 +27,11 @@ class command:
             == self.env["input"]["new_caps_lock"]
         ):
             return
+        current_input = self.env["input"]["curr_input"]
+        previous_input = self.env["input"]["prev_input"]
+        relevant_input = current_input or previous_input
+        if "KEY_CAPSLOCK" not in relevant_input:
+            return
         if self.env["input"]["new_caps_lock"]:
             self.env["runtime"]["OutputManager"].present_text(
                 _("Capslock on"), interrupt=True

src/fenrirscreenreader/commands/onKeyInput/80300-scrolllock.py

@@ -27,6 +27,11 @@ class command:
             == self.env["input"]["new_scroll_lock"]
         ):
             return
+        current_input = self.env["input"]["curr_input"]
+        previous_input = self.env["input"]["prev_input"]
+        relevant_input = current_input or previous_input
+        if "KEY_SCROLLLOCK" not in relevant_input:
+            return
         if self.env["input"]["new_scroll_lock"]:
             self.env["runtime"]["OutputManager"].present_text(
                 _("Scrolllock on"), interrupt=True

src/fenrirscreenreader/commands/onKeyInput/80500-numlock.py

@@ -28,15 +28,21 @@ class command:
         # Only announce numlock changes if an actual numlock key was pressed
         # AND the LED state actually changed (some numpads send spurious NUMLOCK events)
         current_input = self.env["input"]["curr_input"]
+        previous_input = self.env["input"]["prev_input"]
+        relevant_input = current_input or previous_input
 
         # Check if this is a genuine numlock key press by verifying:
         # 1. KEY_NUMLOCK is in the current input sequence
         # 2. The LED state has actually changed
         # 3. This isn't just a side effect from a KP_ key (which some buggy numpads do)
         is_genuine_numlock = (
-            current_input and
-            "KEY_NUMLOCK" in current_input and
-            not any(key.startswith("KEY_KP") for key in current_input if isinstance(key, str))
+            relevant_input and
+            "KEY_NUMLOCK" in relevant_input and
+            not any(
+                key.startswith("KEY_KP")
+                for key in relevant_input
+                if isinstance(key, str)
+            )
         )
 
         if is_genuine_numlock:

src/fenrirscreenreader/commands/onScreenUpdate/60000-history.py

@@ -41,20 +41,11 @@ class command:
             == self.env["runtime"]["ScreenManager"].get_rows() - 1
         ):
             return
-        if self.env["runtime"]["InputManager"].get_shortcut_type() in ["KEY"]:
-            if not (
-                self.env["runtime"]["InputManager"].get_last_deepest_input()
-                in [["KEY_UP"], ["KEY_DOWN"]]
-            ):
-                return
-        elif self.env["runtime"]["InputManager"].get_shortcut_type() in [
-            "BYTE"
-        ]:
-            if not (
-                self.env["runtime"]["ByteManager"].get_last_byte_key()
-                in [b"^[[A", b"^[[B"]
-            ):
-                return
+        if not (
+            self.env["runtime"]["InputManager"].get_last_deepest_input()
+            in [["KEY_UP"], ["KEY_DOWN"]]
+        ):
+            return
 
         # Get the current cursor's line from both old and new content
         prev_line = self.env["screen"]["old_content_text"].split("\n")[

src/fenrirscreenreader/commands/onScreenUpdate/64000-tab_completion.py

@@ -0,0 +1,44 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+
+# Fenrir TTY screen reader
+# By Chrys, Storm Dragon, and contributors.
+
+from fenrirscreenreader.core.i18n import _
+from fenrirscreenreader.core.tabCompletionManager import TabCompletionManager
+
+
+class command:
+    def __init__(self):
+        self.manager = TabCompletionManager()
+
+    def initialize(self, environment):
+        self.env = environment
+        self.manager.initialize(environment)
+
+    def shutdown(self):
+        pass
+
+    def get_description(self):
+        return _("Announces visible tab completion results")
+
+    def run(self):
+        text = self.manager.process_update()
+        if not text:
+            return
+
+        do_interrupt = True
+        if self.env["runtime"]["SettingsManager"].get_setting_as_bool(
+            "speech", "auto_read_incoming"
+        ):
+            do_interrupt = False
+
+        self.env["runtime"]["OutputManager"].present_text(
+            text,
+            interrupt=do_interrupt,
+            announce_capital=True,
+            flush=False,
+        )
+
+    def set_callback(self, callback):
+        pass

src/fenrirscreenreader/commands/onScreenUpdate/65000-progress_detector.py

@@ -36,6 +36,7 @@ class command:
                 # screen changes
                 elif (
                     self.env["screen"]["new_delta"]
+                    and not self.env["screen"].get("new_delta_is_typing", False)
                     and self.is_real_progress_update()
                 ):
                     self.detect_progress(self.env["screen"]["new_delta"])
@@ -78,18 +79,24 @@ class command:
         # updates
         delta_text = self.env["screen"]["new_delta"]
         delta_length = len(delta_text)
+        explicit_progress_lines = self.get_explicit_progress_lines(delta_text)
         if (
             delta_length > 200
-        ):  # Allow longer progress lines like Claude Code's status
-            self.env["runtime"]["DebugManager"].write_debug_out(
-                f"Progress filter: delta too long ({delta_length})",
-                debug.DebugLevel.INFO,
-            )
-            return False
+        ):  # Allow longer progress lines such as terminal status output
+            if not explicit_progress_lines:
+                self.env["runtime"]["DebugManager"].write_debug_out(
+                    f"Progress filter: delta too long ({delta_length})",
+                    debug.DebugLevel.INFO,
+                )
+                return False
 
         # If delta contains newlines and is substantial, let incoming handler
         # deal with it to avoid interfering with multi-line text output
-        if '\n' in delta_text and delta_length > 50:
+        if (
+            "\n" in delta_text
+            and delta_length > 50
+            and not explicit_progress_lines
+        ):
             self.env["runtime"]["DebugManager"].write_debug_out(
                 f"Progress filter: multiline delta ({delta_length} chars)",
                 debug.DebugLevel.INFO,
@@ -106,6 +113,84 @@ class command:
 
         return True
 
+    def get_curl_classic_percentage(self, text):
+        """Return progress from a structurally valid curl classic row."""
+        import re
+
+        size_field = r"\d+(?:\.\d+)?(?:[kKmMgGtT](?:i?[bB])?)?"
+        time_field = r"(?:(?:\d+:)?\d{2}:\d{2}|--:--:--)"
+        curl_classic_match = re.match(
+            rf"^\s*(\d+)\s+{size_field}\s+"
+            rf"(\d+)\s+{size_field}\s+"
+            rf"(\d+)\s+{size_field}\s+"
+            rf"{size_field}\s+{size_field}\s+"
+            rf"{time_field}\s+{time_field}\s+{time_field}\s+"
+            rf"{size_field}(?=\s|$)",
+            text,
+        )
+        if not curl_classic_match:
+            return None
+
+        total_percentage = int(curl_classic_match.group(1))
+        received_percentage = int(curl_classic_match.group(2))
+        uploaded_percentage = int(curl_classic_match.group(3))
+        if not all(
+            0 <= percentage <= 100
+            for percentage in (
+                total_percentage,
+                received_percentage,
+                uploaded_percentage,
+            )
+        ):
+            return None
+        if total_percentage != received_percentage:
+            return None
+        return float(total_percentage)
+
+    def is_explicit_progress_delta(self, text):
+        """Allow long single-line deltas that still look like progress output."""
+        import re
+
+        if "\n" in text or self.contains_url(text):
+            return False
+
+        if self.get_curl_classic_percentage(text) is not None:
+            return True
+
+        has_percentage = re.search(r"(^|\s)\d+(?:\.\d+)?\s*%", text)
+        if not has_percentage:
+            return self.has_progress_status_fields(text)
+
+        bracketed_bar = re.search(
+            r"\[(?=[#=*>█▉▊▋▌▍▎▏▒▓░.\s-]{2,}\])"
+            r"(?=[^\]]*[#=*>█▉▊▋▌▍▎▏▒▓░])"
+            r"[#=*>█▉▊▋▌▍▎▏▒▓░.\s-]+\]",
+            text,
+        )
+        pipe_bar = re.search(
+            r"\|(?=[#=*>█▉▊▋▌▍▎▏▒▓░.\s-]{2,}\|)"
+            r"(?=[^|]*[#=*>█▉▊▋▌▍▎▏▒▓░])"
+            r"[#=*>█▉▊▋▌▍▎▏▒▓░.\s-]+\|",
+            text,
+        )
+        transfer_rate = re.search(
+            r"\b\d+(?:\.\d+)?\s*[kKmMgGtT](?:i?B)?/s\b",
+            text,
+        )
+        return bool(
+            bracketed_bar
+            or pipe_bar
+            or transfer_rate
+        )
+
+    def get_explicit_progress_lines(self, text):
+        """Return structurally explicit progress rows from a screen delta."""
+        return [
+            line
+            for line in text.splitlines()
+            if self.is_explicit_progress_delta(line)
+        ]
+
     def reset_progress_state(self):
         """Reset progress state when a prompt is detected, allowing new progress operations to start fresh"""
         self.env["runtime"]["DebugManager"].write_debug_out(
@@ -119,6 +204,15 @@ class command:
         import re
         import time
 
+        if "\n" in text:
+            explicit_progress_lines = self.get_explicit_progress_lines(text)
+            if not explicit_progress_lines:
+                return
+            # Parallel transfer tools commonly repaint several package rows,
+            # followed by an aggregate row. One tone per update is enough, and
+            # the final explicit row is normally that aggregate.
+            text = explicit_progress_lines[-1]
+
         current_time = time.time()
 
         # Debug: Print what we're checking
@@ -172,34 +266,21 @@ class command:
                     return
 
         # Pattern 1a2: Curl classic progress format (percentage without % symbol)
-        # Extract percentage from curl's classic format
-        curl_classic_match = re.search(
-            r"^\s*(\d+)\s+\d+[kMGT]?\s+(\d+)\s+\d+[kMGT]?\s+\d+\s+\d+\s+\d+[kMGT]?\s+\d+\s+\d+:\d+:\d+\s+\d+:\d+:\d+\s+\d+:\d+:\d+\s+\d+[kMGT]?\s*$", text
-        )
-        if curl_classic_match:
-            # Use the first percentage (total progress)
-            percentage = float(curl_classic_match.group(1))
-            if 0 <= percentage <= 100:
+        percentage = self.get_curl_classic_percentage(text)
+        if percentage is not None:
+            self.env["runtime"]["DebugManager"].write_debug_out(
+                "found curl classic percentage: " + str(percentage),
+                debug.DebugLevel.INFO,
+            )
+            if percentage != self.env["commandBuffer"]["lastProgressValue"]:
                 self.env["runtime"]["DebugManager"].write_debug_out(
-                    "found curl classic percentage: " + str(percentage),
+                    "Playing tone for curl: " + str(percentage),
                     debug.DebugLevel.INFO,
                 )
-                if (
-                    percentage
-                    != self.env["commandBuffer"]["lastProgressValue"]
-                ):
-                    self.env["runtime"]["DebugManager"].write_debug_out(
-                        "Playing tone for curl: " + str(percentage),
-                        debug.DebugLevel.INFO,
-                    )
-                    self.play_progress_tone(percentage)
-                    self.env["commandBuffer"][
-                        "lastProgressValue"
-                    ] = percentage
-                    self.env["commandBuffer"][
-                        "lastProgressTime"
-                    ] = current_time
-                return
+                self.play_progress_tone(percentage)
+                self.env["commandBuffer"]["lastProgressValue"] = percentage
+                self.env["commandBuffer"]["lastProgressTime"] = current_time
+            return
 
         # Pattern 1b: Time/token activity (not percentage-based, so use single
         # beep)
@@ -219,8 +300,20 @@ class command:
         pacman_match = re.search(
             r"\d+(?:\.\d+)?\s+[kKmMgGtT]iB\s+\d+(?:\.\d+)?\s+[kKmMgGtT]iB/s\s+\d+:\d+", text
         )
+        # Pattern 1g: Key/value status lines from ffmpeg-like terminal tools.
+        # These do not expose a percentage in the status line, so treat them as
+        # activity rather than guessing progress from elapsed time.
+        status_fields_match = self.has_progress_status_fields(text)
 
-        if time_match or token_match or dd_match or curl_match or transfer_match or pacman_match:
+        if (
+            time_match
+            or token_match
+            or dd_match
+            or curl_match
+            or transfer_match
+            or pacman_match
+            or status_fields_match
+        ):
             # For non-percentage progress, use a single activity beep every 2
             # seconds
             if (
@@ -305,43 +398,41 @@ class command:
                 self.env["commandBuffer"]["lastProgressTime"] = current_time
             return
 
-        # Pattern 6: Claude Code working indicators (various symbols + activity text + "esc/ctrl+c to interrupt")
-        # Matches any: [symbol] [Task description]… (... to interrupt ...)
+        # Pattern 6: Interruptible terminal activity indicators
+        # Matches any: [symbol] [Task description][…] (... to interrupt ...)
         # Symbols include: * ✢ ✽ ✶ ✻ · • ◦ ○ ● ◆ and similar decorative characters
-        # Example: ✽ Reviewing script for issues… (ctrl+c to interrupt · 33s · ↑ 1.6k tokens · thought for 4s)
-        claude_progress_match = re.search(
-            r'[*✢✽✶✻·•◦○●◆]\s+\w+.*?…\s*\(.*(?:esc|ctrl\+c) to interrupt.*\)',
+        # Keep this structural rather than adding application-specific formats,
+        # which change too frequently to support reliably.
+        interruptible_activity_match = re.search(
+            r'[*✢✽✶✻·•◦○●◆]\s+\w+.*?(?:…\s*)?\(.*(?:esc|ctrl\+c) to interrupt.*\)',
             text,
             re.IGNORECASE,
         )
-        if claude_progress_match:
+        if not interruptible_activity_match:
+            interruptible_activity_match = re.search(
+                (
+                    r'^\s*(?:thinking|working|processing|analyzing|searching|'
+                    r'reading|writing|planning|running|executing|updating|'
+                    r'building|installing|compiling|downloading|reviewing|'
+                    r'generating|responding|applying|fixing|editing|creating|'
+                    r'preparing|checking|opening|loading|fetching|retrieving|'
+                    r'scanning|indexing|summarizing)\b.*'
+                    r'\([^)]*(?:esc|ctrl\+c) to interrupt[^)]*\)\s*$'
+                ),
+                text,
+                re.IGNORECASE,
+            )
+        if interruptible_activity_match:
             if current_time - self.env["commandBuffer"]["lastProgressTime"] >= 1.0:
                 self.env["runtime"]["DebugManager"].write_debug_out(
-                    "Playing Claude Code activity beep",
+                    "Playing interruptible activity beep",
                     debug.DebugLevel.INFO,
                 )
                 self.play_activity_beep()
                 self.env["commandBuffer"]["lastProgressTime"] = current_time
             return
 
-        # Pattern 6b: Claude Code tool invocation indicators (● Tool Name(...))
-        # Example: ● Web Search("query here")
-        tool_invocation_match = re.search(
-            r'[●○◉•◦]\s+(?:Web\s*Search|Read|Write|Edit|Bash|Glob|Grep|Task|WebFetch)\s*\(',
-            text,
-            re.IGNORECASE,
-        )
-        if tool_invocation_match:
-            if current_time - self.env["commandBuffer"]["lastProgressTime"] >= 1.0:
-                self.env["runtime"]["DebugManager"].write_debug_out(
-                    "Playing Claude Code tool invocation beep",
-                    debug.DebugLevel.INFO,
-                )
-                self.play_activity_beep()
-                self.env["commandBuffer"]["lastProgressTime"] = current_time
-            return
-
-        # Pattern 6c: Bullet/white bullet activity lines (•/◦ ...)
+        # Pattern 6b: Bullet/white bullet activity lines (•/◦ ...)
         bullet_activity_match = re.search(
             (
                 r'^\s*[•◦]\s+.*(?:…|\.{3,}|\b(?:thinking|working|processing|'
@@ -572,5 +663,48 @@ class command:
                 return True
         return False
 
+    def has_progress_status_fields(self, text):
+        """Recognize dense key=value status lines from terminal tools."""
+        import re
+
+        if "\n" in text or self.contains_url(text):
+            return False
+
+        fields = re.findall(r"\b([a-zA-Z][a-zA-Z0-9_-]*)=\s*\S+", text)
+        if len(fields) < 3:
+            return False
+
+        field_names = {field.lower() for field in fields}
+        progress_fields = {
+            "frame",
+            "fps",
+            "size",
+            "time",
+            "bitrate",
+            "speed",
+            "elapsed",
+            "dup",
+            "drop",
+        }
+        if len(field_names & progress_fields) < 3:
+            return False
+
+        has_time_field = bool(
+            re.search(
+                r"\b(?:time|elapsed|eta)=\s*\d+(?::\d{2}){1,2}(?:\.\d+)?\b",
+                text,
+                re.IGNORECASE,
+            )
+        )
+        has_rate_field = bool(
+            re.search(
+                r"\b(?:speed|fps|bitrate)=\s*[\d.]+[a-zA-Z/]*\b",
+                text,
+                re.IGNORECASE,
+            )
+        )
+
+        return has_time_field and has_rate_field
+
     def set_callback(self, callback):
         pass

src/fenrirscreenreader/commands/onScreenUpdate/70000-incoming.py

@@ -4,6 +4,8 @@
 # Fenrir TTY screen reader
 # By Chrys, Storm Dragon, and contributors.
 
+import difflib
+import re
 import time
 from fenrirscreenreader.core.i18n import _
 
@@ -15,6 +17,8 @@ class command:
         self._batched_text = []
         self._last_flush_time = 0
         self._in_flood_mode = False
+        self._pending_focus_text = ""
+        self._pending_focus_time = 0
 
     def initialize(self, environment):
         self.env = environment
@@ -111,6 +115,131 @@ class command:
 
         return False
 
+    def _normalize_line(self, line):
+        return " ".join(line.split())
+
+    def _is_focus_control_line(self, text):
+        stripped = self._normalize_line(text)
+        if not stripped:
+            return False
+        if re.fullmatch(r"<\s*[A-Za-z][A-Za-z0-9 _.-]{0,30}\s*>", stripped):
+            return True
+        return stripped.lower() in {
+            "ok",
+            "cancel",
+            "yes",
+            "no",
+            "retry",
+            "abort",
+            "ignore",
+            "continue",
+        }
+
+    def _is_transient_focus_only_update(self, delta_text):
+        delta_lines = [
+            self._normalize_line(line)
+            for line in delta_text.splitlines()
+            if line.strip()
+        ]
+        if len(delta_lines) != 1:
+            return False
+        if not self._is_focus_control_line(delta_lines[0]):
+            return False
+
+        screen_lines = [
+            self._normalize_line(line)
+            for line in self.env["screen"]["new_content_text"].split("\n")
+            if line.strip()
+        ]
+        return screen_lines == delta_lines
+
+    def _store_pending_focus_text(self, delta_text):
+        delta_lines = [
+            self._normalize_line(line)
+            for line in delta_text.splitlines()
+            if line.strip()
+        ]
+        self._pending_focus_text = "\n".join(delta_lines)
+        self._pending_focus_time = time.time()
+
+    def _take_pending_focus_text(self):
+        if not self._pending_focus_text:
+            return ""
+        if time.time() - self._pending_focus_time > 0.5:
+            self._pending_focus_text = ""
+            self._pending_focus_time = 0
+            return ""
+        pending = self._pending_focus_text
+        self._pending_focus_text = ""
+        self._pending_focus_time = 0
+        return pending
+
+    def _is_subsequence(self, subset_lines, source_lines):
+        subset_index = 0
+        for source_line in source_lines:
+            if source_line == subset_lines[subset_index]:
+                subset_index += 1
+                if subset_index == len(subset_lines):
+                    return True
+        return False
+
+    def _prefer_inserted_lower_screen_text(self, delta_text):
+        delta_lines = [
+            line for line in delta_text.splitlines() if line.strip() != ""
+        ]
+        if len(delta_lines) < 2:
+            return delta_text
+
+        old_lines = self.env["screen"]["old_content_text"].split("\n")
+        new_lines = self.env["screen"]["new_content_text"].split("\n")
+        if len(old_lines) != len(new_lines) or len(new_lines) < 4:
+            return delta_text
+
+        normalized_old_lines = [
+            self._normalize_line(line) for line in old_lines
+        ]
+        normalized_new_lines = [
+            self._normalize_line(line) for line in new_lines
+        ]
+        matcher = difflib.SequenceMatcher(
+            None, normalized_old_lines, normalized_new_lines, autojunk=False
+        )
+
+        top_lines_changed = False
+        inserted_lines = []
+        lower_screen_start = max(2, len(new_lines) // 2)
+
+        for tag, old_start, old_end, new_start, new_end in matcher.get_opcodes():
+            if tag in {"replace", "delete"} and old_start < 2:
+                top_lines_changed = True
+            if tag != "insert" or new_start < lower_screen_start:
+                continue
+            inserted_lines.extend(
+                line for line in new_lines[new_start:new_end] if line.strip() != ""
+            )
+
+        if not top_lines_changed or not inserted_lines:
+            return delta_text
+        if all(
+            self._is_focus_control_line(line) for line in inserted_lines
+        ):
+            return delta_text
+
+        normalized_delta_lines = [
+            self._normalize_line(line) for line in delta_lines
+        ]
+        normalized_inserted_lines = [
+            self._normalize_line(line) for line in inserted_lines
+        ]
+        if not self._is_subsequence(
+            normalized_inserted_lines, normalized_delta_lines
+        ):
+            return delta_text
+        if normalized_delta_lines == normalized_inserted_lines:
+            return delta_text
+
+        return "\n".join(inserted_lines)
+
     def run(self):
         if not self.env["runtime"]["SettingsManager"].get_setting_as_bool(
             "speech", "auto_read_incoming"
@@ -119,11 +248,15 @@ class command:
         # is there something to read?
         if not self.env["runtime"]["ScreenManager"].is_delta(ignoreSpace=True):
             return
+        if self.env["screen"].get("new_delta_is_typing", False):
+            return
 
         delta_text = self.env["screen"]["new_delta"]
 
         if self.env["runtime"]["ScreenManager"].is_screen_change():
             self._reset_flood_state()
+            self._pending_focus_text = ""
+            self._pending_focus_time = 0
         
         # Skip if tab completion already handled this delta
         if self._was_handled_by_tab_completion(delta_text):
@@ -146,6 +279,10 @@ class command:
             if "\n" not in delta_text:
                 return
 
+        if self._is_transient_focus_only_update(delta_text):
+            self._store_pending_focus_text(delta_text)
+            return
+
         rapid = self._is_rapid_updates()
         high_volume = self._is_high_volume(delta_text)
 
@@ -169,6 +306,10 @@ class command:
             return
 
         # print(x_move, y_move, len(self.env['screen']['new_delta']), len(self.env['screen']['newNegativeDelta']))
+        delta_text = self._prefer_inserted_lower_screen_text(delta_text)
+        pending_focus_text = self._take_pending_focus_text()
+        if pending_focus_text:
+            delta_text = "\n".join([delta_text, pending_focus_text])
         self.env["runtime"]["OutputManager"].present_text(
             delta_text, interrupt=False, flush=False
         )

src/fenrirscreenreader/commands/speech-history/speech_history_close.py

@@ -0,0 +1,27 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+
+# Fenrir TTY screen reader
+# By Chrys, Storm Dragon, and contributors.
+
+from fenrirscreenreader.core.i18n import _
+
+
+class command:
+    def __init__(self):
+        pass
+
+    def initialize(self, environment):
+        self.env = environment
+
+    def shutdown(self):
+        pass
+
+    def get_description(self):
+        return _("closes speech history")
+
+    def run(self):
+        self.env["runtime"]["SpeechHistoryManager"].close_history()
+
+    def set_callback(self, callback):
+        pass

src/fenrirscreenreader/commands/speech-history/speech_history_copy.py

@@ -0,0 +1,27 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+
+# Fenrir TTY screen reader
+# By Chrys, Storm Dragon, and contributors.
+
+from fenrirscreenreader.core.i18n import _
+
+
+class command:
+    def __init__(self):
+        pass
+
+    def initialize(self, environment):
+        self.env = environment
+
+    def shutdown(self):
+        pass
+
+    def get_description(self):
+        return _("copies current speech history item to the clipboard")
+
+    def run(self):
+        self.env["runtime"]["SpeechHistoryManager"].copy_current_to_clipboard()
+
+    def set_callback(self, callback):
+        pass

src/fenrirscreenreader/commands/speech-history/speech_history_current.py

@@ -0,0 +1,27 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+
+# Fenrir TTY screen reader
+# By Chrys, Storm Dragon, and contributors.
+
+from fenrirscreenreader.core.i18n import _
+
+
+class command:
+    def __init__(self):
+        pass
+
+    def initialize(self, environment):
+        self.env = environment
+
+    def shutdown(self):
+        pass
+
+    def get_description(self):
+        return _("speaks current speech history item")
+
+    def run(self):
+        self.env["runtime"]["SpeechHistoryManager"].present_current()
+
+    def set_callback(self, callback):
+        pass

src/fenrirscreenreader/commands/speech-history/speech_history_next.py

@@ -0,0 +1,27 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+
+# Fenrir TTY screen reader
+# By Chrys, Storm Dragon, and contributors.
+
+from fenrirscreenreader.core.i18n import _
+
+
+class command:
+    def __init__(self):
+        pass
+
+    def initialize(self, environment):
+        self.env = environment
+
+    def shutdown(self):
+        pass
+
+    def get_description(self):
+        return _("selects the next speech history item")
+
+    def run(self):
+        self.env["runtime"]["SpeechHistoryManager"].next_entry()
+
+    def set_callback(self, callback):
+        pass

src/fenrirscreenreader/commands/speech-history/speech_history_prev.py

@@ -0,0 +1,27 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+
+# Fenrir TTY screen reader
+# By Chrys, Storm Dragon, and contributors.
+
+from fenrirscreenreader.core.i18n import _
+
+
+class command:
+    def __init__(self):
+        pass
+
+    def initialize(self, environment):
+        self.env = environment
+
+    def shutdown(self):
+        pass
+
+    def get_description(self):
+        return _("selects the previous speech history item")
+
+    def run(self):
+        self.env["runtime"]["SpeechHistoryManager"].prev_entry()
+
+    def set_callback(self, callback):
+        pass

src/fenrirscreenreader/commands/vmenu-profiles/KEY/fenrir/config_base.py

@@ -127,6 +127,8 @@ class config_command:
             self.config.set("speech", "rate", "0.75")
             self.config.set("speech", "pitch", "0.5")
             self.config.set("speech", "volume", "1.0")
+            self.config.set("speech", "hardware_device", "/dev/ttyS0")
+            self.config.set("speech", "hardware_baud_rate", "9600")
 
             self.config.add_section("sound")
             self.config.set("sound", "driver", "genericDriver")

src/fenrirscreenreader/commands/vmenu-profiles/KEY/fenrir/management/reset_defaults.py

@@ -108,6 +108,8 @@ class command(config_command):
                 "rate": "0.5",
                 "pitch": "0.5",
                 "volume": "1.0",
+                "hardware_device": "/dev/ttyS0",
+                "hardware_baud_rate": "9600",
                 "auto_read_incoming": "True",
             }
 
@@ -139,7 +141,12 @@ class command(config_command):
             self.config["general"] = {
                 "punctuation_level": "some",
                 "debug_level": "0",
+            }
+
+            # Basic clipboard defaults
+            self.config["clipboard"] = {
                 "number_of_clipboards": "50",
+                "clipboard_export_path": "/tmp/fenrirClipboard",
             }
 
             # Write the configuration

src/fenrirscreenreader/commands/vmenu-profiles/KEY/fenrir/screen/select_driver.py

@@ -37,7 +37,7 @@ class command(config_command):
         self.present_text(f"Current screen driver: {current_description}")
 
         # Cycle through the available drivers
-        drivers = ["vcsaDriver", "ptyDriver", "dummyDriver", "debugDriver"]
+        drivers = ["vcsaDriver", "ptyDriver", "dummyDriver"]
         try:
             current_index = drivers.index(current_driver)
             next_index = (current_index + 1) % len(drivers)

src/fenrirscreenreader/commands/vmenu-profiles/KEY/fenrir/sound/select_theme.py

@@ -22,23 +22,24 @@ class command(config_command):
 
     def run(self):
         current_theme = self.get_setting("sound", "theme", "default")
+        current_theme_name = os.path.basename(
+            os.path.normpath(current_theme)
+        )
 
         # Present current theme
         self.present_text(f"Current sound theme: {current_theme}")
 
         # Look for available sound themes
-        sound_paths = [
-            "/usr/share/sounds",
-            "/usr/share/fenrirscreenreader/sounds",
-            os.path.expanduser("~/.local/share/fenrirscreenreader/sounds"),
-        ]
+        sound_paths = self.env[
+            "runtime"
+        ]["SettingsManager"].get_sound_theme_roots()
 
         available_themes = self.get_available_themes(sound_paths)
 
         if len(available_themes) > 1:
             # For this implementation, cycle through available themes
             try:
-                current_index = available_themes.index(current_theme)
+                current_index = available_themes.index(current_theme_name)
                 next_index = (current_index + 1) % len(available_themes)
                 new_theme = available_themes[next_index]
             except ValueError:

src/fenrirscreenreader/commands/vmenu-profiles/KEY/nano/Help/about_nano.py

@@ -29,12 +29,7 @@ class command:
         self.env["runtime"]["OutputManager"].present_text(
             "Okay, loading the information about Nano.", interrupt=True
         )
-        if self.env["runtime"]["InputManager"].get_shortcut_type() in ["KEY"]:
-            self.env["runtime"]["InputManager"].send_keys(self.key_macro)
-        elif self.env["runtime"]["InputManager"].get_shortcut_type() in [
-            "BYTE"
-        ]:
-            self.env["runtime"]["ByteManager"].send_bytes(self.byteMakro)
+        self.env["runtime"]["InputManager"].send_keys(self.key_macro)
 
     def set_callback(self, callback):
         pass

src/fenrirscreenreader/commands/vmenu-profiles/KEY/nano/file/save.py

@@ -29,12 +29,7 @@ class command:
         self.env["runtime"]["OutputManager"].present_text(
             "Okay, you will now be asked to save your work.", interrupt=True
         )
-        if self.env["runtime"]["InputManager"].get_shortcut_type() in ["KEY"]:
-            self.env["runtime"]["InputManager"].send_keys(self.key_macro)
-        elif self.env["runtime"]["InputManager"].get_shortcut_type() in [
-            "BYTE"
-        ]:
-            self.env["runtime"]["ByteManager"].send_bytes(self.byteMakro)
+        self.env["runtime"]["InputManager"].send_keys(self.key_macro)
 
     def set_callback(self, callback):
         pass

src/fenrirscreenreader/commands/vmenu-profiles/template.py

@@ -18,7 +18,6 @@ class command:
         # self.key_macro = [[1,'KEY_LEFTCTRL'],[1,'KEY_O'],[0.05,'SLEEP'],[0,'KEY_O'],[0,'KEY_LEFTCTRL']]
         # self.key_macro = [[1,'KEY_LEFTSHIFT'],[1,'KEY_LEFTCTRL'],[1,'KEY_N'],[0.05,'SLEEP'],[0,'KEY_N'],[0,'KEY_LEFTCTRL'],[0,'KEY_LEFTSHIFT']]
         self.key_macro = []
-        self.byteMakro = []
 
     def shutdown(self):
         pass
@@ -27,12 +26,7 @@ class command:
         return "No description found"
 
     def run(self):
-        if self.env["runtime"]["InputManager"].get_shortcut_type() in ["KEY"]:
-            self.env["runtime"]["InputManager"].send_keys(self.key_macro)
-        elif self.env["runtime"]["InputManager"].get_shortcut_type() in [
-            "BYTE"
-        ]:
-            self.env["runtime"]["ByteManager"].send_bytes(self.byteMakro)
+        self.env["runtime"]["InputManager"].send_keys(self.key_macro)
 
     def set_callback(self, callback):
         pass