Sound manager added. Pitch controls for menu voice bound to { and }.

.gitignore

@@ -3,5 +3,5 @@ image/*
 *.img
 *.img.xz
 
-# Claude AI assistant files
-CLAUDE.md
+# AI assistant files
+AGENTS.md

etc/speech-dispatcher/speechd.conf.bak

@@ -288,6 +288,7 @@ AddModule "piper-tts-generic"        "sd_generic"   "piper-tts-generic.conf"
 #AddModule "mary-generic"             "sd_generic"   "mary-generic.conf"
 #AddModule "baratinoo"                "sd_baratinoo" "baratinoo.conf"
 AddModule "rhvoice"                  "sd_rhvoice"   "rhvoice.conf"
+AddModule "viavoice"                 "sd_viavoice"  "viavoice.conf"
 AddModule "voxin"                    "sd_voxin"     "voxin.conf"
 
 # The output module testing doesn't actually connect to anything. It

usr/local/bin/audio_manager.py

@@ -0,0 +1,600 @@
+#!/usr/bin/env python3
+
+import curses
+import json
+import os
+import re
+import shutil
+import subprocess
+import sys
+import time
+from dataclasses import dataclass
+
+try:
+    import simpleaudio as sa
+except Exception:
+    sa = None
+
+try:
+    import speechd
+except Exception:
+    speechd = None
+
+
+FEEDBACK_SOUND = "/usr/share/sounds/stormux/menu_move.wav"
+TEST_SOUND = "/usr/share/sounds/stormux/menu_select.wav"
+HELP_TEXT = (
+    "Use left and right to move between controls. "
+    "Use up and down to change the selected control. "
+    "Press space to hear the current value. "
+    "Press enter to confirm a pending device change or apply the selected device. "
+    "Press escape or q to exit."
+)
+
+
+def clamp_percent(value, minimum, maximum):
+    return max(minimum, min(maximum, value))
+
+
+def format_percent_text(value):
+    return f"{value} percent"
+
+
+def cycle_index(current_index, delta, total_count):
+    if total_count <= 0:
+        return 0
+    return (current_index + delta) % total_count
+
+
+def pending_change_expired(change, now_value=None):
+    if change is None:
+        return False
+    if now_value is None:
+        now_value = time.time()
+    return (now_value - change.started_at) >= change.timeout_seconds
+
+
+def build_default_command(device_kind, device_name):
+    if device_kind == "sink":
+        return ["pactl", "set-default-sink", str(device_name)]
+    if device_kind == "source":
+        return ["pactl", "set-default-source", str(device_name)]
+    raise ValueError(f"Unsupported device kind: {device_kind}")
+
+
+def build_volume_command(device_kind, device_name, volume_percent):
+    bounded = clamp_percent(volume_percent, 0, 150)
+    if device_kind == "sink":
+        return ["pactl", "set-sink-volume", str(device_name), f"{bounded}%"]
+    if device_kind == "source":
+        return ["pactl", "set-source-volume", str(device_name), f"{bounded}%"]
+    raise ValueError(f"Unsupported device kind: {device_kind}")
+
+
+def is_monitor_source(device_name, description):
+    lowered_name = (device_name or "").lower()
+    lowered_description = (description or "").lower()
+    return lowered_name.endswith(".monitor") or lowered_description.startswith("monitor of ")
+
+
+def extract_devices(payload, exclude_monitors=False):
+    devices = []
+    for item in payload:
+        device_id = item.get("index")
+        node_name = item.get("name", "")
+        properties = item.get("properties") or {}
+        description = (
+            properties.get("device.description")
+            or item.get("description")
+            or properties.get("node.description")
+            or node_name
+            or f"Device {device_id}"
+        )
+
+        if device_id is None:
+            continue
+
+        if exclude_monitors and is_monitor_source(node_name, description):
+            continue
+
+        devices.append(
+            {
+                "id": device_id,
+                "name": description,
+                "node_name": node_name,
+            }
+        )
+    return devices
+
+
+def parse_percent_volume(output_text):
+    match = re.search(r"(\d+)%", output_text)
+    if not match:
+        raise RuntimeError(f"Unexpected percent volume output: {output_text}")
+    return int(match.group(1))
+
+
+def parse_fraction_volume(output_text):
+    match = re.search(r"([0-9]*\.?[0-9]+)", output_text)
+    if not match:
+        raise RuntimeError(f"Unexpected fractional volume output: {output_text}")
+    return round(float(match.group(1)) * 100)
+
+
+def build_external_sound_command(file_path):
+    player_candidates = [
+        ("pw-play", ["pw-play", file_path]),
+        ("paplay", ["paplay", file_path]),
+        ("aplay", ["aplay", file_path]),
+        ("play", ["play", "-q", file_path]),
+    ]
+    for executable, command in player_candidates:
+        if shutil.which(executable) is not None:
+            return command
+    return None
+
+
+def build_feedback_tone_command():
+    if shutil.which("sox") is None:
+        return None
+    return ["sox", "-nqdV0", "synth", ".1", "tri", "840", "fade", ".04", ".1", ".04"]
+
+
+def find_device_index(devices, *, node_name=None, device_id=None):
+    for index, device in enumerate(devices):
+        if device_id is not None and device["id"] == device_id:
+            return index
+        if node_name is not None and device["node_name"] == node_name:
+            return index
+    return 0
+
+
+@dataclass
+class PendingDeviceChange:
+    device_kind: str
+    previous_target: str
+    candidate_target: str
+    candidate_name: str
+    started_at: float = 0.0
+    timeout_seconds: int = 15
+
+
+class AudioBackend:
+    def run_command(self, command):
+        return subprocess.run(command, capture_output=True, text=True, check=True)
+
+    def list_sinks(self):
+        payload = json.loads(self.run_command(["pactl", "--format=json", "list", "sinks"]).stdout)
+        return extract_devices(payload)
+
+    def list_sources(self):
+        payload = json.loads(self.run_command(["pactl", "--format=json", "list", "sources"]).stdout)
+        return extract_devices(payload, exclude_monitors=True)
+
+    def get_default_sink_name(self):
+        return self.run_command(["pactl", "get-default-sink"]).stdout.strip()
+
+    def get_default_source_name(self):
+        return self.run_command(["pactl", "get-default-source"]).stdout.strip()
+
+    def set_default_device(self, device_kind, device_name):
+        self.run_command(build_default_command(device_kind, device_name))
+
+    def set_device_volume(self, device_kind, device_name, percent_value):
+        self.run_command(build_volume_command(device_kind, device_name, percent_value))
+
+    def get_device_volume(self, device_kind, device_name):
+        if device_kind == "sink":
+            output = self.run_command(["pactl", "get-sink-volume", str(device_name)]).stdout.strip()
+        elif device_kind == "source":
+            output = self.run_command(["pactl", "get-source-volume", str(device_name)]).stdout.strip()
+        else:
+            raise ValueError(f"Unsupported device kind: {device_kind}")
+        return parse_percent_volume(output)
+
+
+class AudioFeedback:
+    def __init__(self):
+        self.current_sound = None
+        self.current_process = None
+
+    def stop_sound(self):
+        if self.current_sound is not None and self.current_sound.is_playing():
+            self.current_sound.stop()
+        self.current_sound = None
+        if self.current_process is not None and self.current_process.poll() is None:
+            self.current_process.terminate()
+        self.current_process = None
+
+    def play_sound_file(self, file_path):
+        if sa is not None and os.path.exists(file_path):
+            try:
+                self.stop_sound()
+                self.current_sound = sa.WaveObject.from_wave_file(file_path).play()
+                return
+            except Exception:
+                pass
+
+        external_command = build_external_sound_command(file_path)
+        if external_command is not None:
+            try:
+                self.stop_sound()
+                self.current_process = subprocess.Popen(
+                    external_command,
+                    stdout=subprocess.DEVNULL,
+                    stderr=subprocess.DEVNULL,
+                )
+                return
+            except Exception:
+                pass
+
+        curses.beep()
+
+    def play_feedback_beep(self):
+        tone_command = build_feedback_tone_command()
+        if tone_command is not None:
+            try:
+                self.stop_sound()
+                self.current_process = subprocess.Popen(
+                    tone_command,
+                    stdout=subprocess.DEVNULL,
+                    stderr=subprocess.DEVNULL,
+                )
+                return
+            except Exception:
+                pass
+
+        self.play_sound_file(FEEDBACK_SOUND)
+
+    def play_test_beep(self):
+        self.play_sound_file(TEST_SOUND)
+
+
+class AudioManagerApp:
+    def __init__(self):
+        self.backend = AudioBackend()
+        self.feedback = AudioFeedback()
+        self.speech_client = None
+        self.stdscr = None
+        self.focus_names = [
+            "Output Volume",
+            "Microphone Volume",
+            "Output Device",
+            "Microphone Device",
+        ]
+        self.focus_index = 0
+        self.output_devices = []
+        self.input_devices = []
+        self.output_index = 0
+        self.input_index = 0
+        self.default_output_id = None
+        self.default_input_id = None
+        self.default_output_name = ""
+        self.default_input_name = ""
+        self.output_volume = 50
+        self.input_volume = 50
+        self.pending_change = None
+        self.status_message = ""
+
+    def init_speech(self):
+        if speechd is None:
+            return
+        self.speech_client = speechd.SSIPClient("audio_manager")
+        self.speech_client.set_priority(speechd.Priority.IMPORTANT)
+        self.speech_client.set_punctuation(speechd.PunctuationMode.SOME)
+
+    def speak(self, text, interrupt=True):
+        if not self.speech_client:
+            return
+
+        try:
+            if interrupt:
+                self.speech_client.cancel()
+            self.speech_client.speak(text)
+        except Exception:
+            pass
+
+    def set_status(self, message, speak_message=False):
+        self.status_message = message
+        if speak_message:
+            self.speak(message)
+
+    def get_selected_output_device(self):
+        if not self.output_devices:
+            return None
+        return self.output_devices[self.output_index]
+
+    def get_selected_input_device(self):
+        if not self.input_devices:
+            return None
+        return self.input_devices[self.input_index]
+
+    def refresh_state(self, preserve_selection=False):
+        previous_output_id = None
+        previous_input_id = None
+        if preserve_selection:
+            selected_output = self.get_selected_output_device()
+            selected_input = self.get_selected_input_device()
+            previous_output_id = None if selected_output is None else selected_output["id"]
+            previous_input_id = None if selected_input is None else selected_input["id"]
+
+        self.output_devices = self.backend.list_sinks()
+        self.input_devices = self.backend.list_sources()
+
+        if not self.output_devices:
+            raise RuntimeError("No output devices found.")
+        if not self.input_devices:
+            raise RuntimeError("No microphone devices found.")
+
+        default_output_name = self.backend.get_default_sink_name()
+        default_input_name = self.backend.get_default_source_name()
+        self.default_output_name = default_output_name
+        self.default_input_name = default_input_name
+
+        self.output_index = find_device_index(
+            self.output_devices,
+            device_id=previous_output_id if preserve_selection else None,
+            node_name=None if preserve_selection else default_output_name,
+        )
+        self.input_index = find_device_index(
+            self.input_devices,
+            device_id=previous_input_id if preserve_selection else None,
+            node_name=None if preserve_selection else default_input_name,
+        )
+
+        self.default_output_id = self.output_devices[
+            find_device_index(self.output_devices, node_name=default_output_name)
+        ]["id"]
+        self.default_input_id = self.input_devices[
+            find_device_index(self.input_devices, node_name=default_input_name)
+        ]["id"]
+
+        self.output_volume = self.backend.get_device_volume("sink", self.default_output_name)
+        self.input_volume = self.backend.get_device_volume("source", self.default_input_name)
+
+    def apply_volume_delta(self, delta):
+        try:
+            if self.focus_index == 0:
+                new_value = clamp_percent(self.output_volume + delta, 0, 150)
+                self.backend.set_device_volume("sink", self.default_output_name, new_value)
+                self.output_volume = new_value
+                self.feedback.play_feedback_beep()
+            elif self.focus_index == 1:
+                new_value = clamp_percent(self.input_volume + delta, 0, 150)
+                self.backend.set_device_volume("source", self.default_input_name, new_value)
+                self.input_volume = new_value
+                self.feedback.play_feedback_beep()
+        except Exception as error:
+            self.set_status(f"Volume change failed: {error}", speak_message=True)
+
+    def announce_current_focus(self):
+        if self.focus_index == 0:
+            self.speak(format_percent_text(self.output_volume))
+        elif self.focus_index == 1:
+            self.speak(format_percent_text(self.input_volume))
+        elif self.focus_index == 2:
+            device = self.get_selected_output_device()
+            if device:
+                self.speak(device["name"])
+        elif self.focus_index == 3:
+            device = self.get_selected_input_device()
+            if device:
+                self.speak(device["name"])
+
+    def move_focus(self, delta):
+        self.focus_index = cycle_index(self.focus_index, delta, len(self.focus_names))
+        self.speak(self.focus_names[self.focus_index])
+
+    def handle_up(self):
+        if self.focus_index in (0, 1):
+            self.apply_volume_delta(5)
+            return
+
+        if self.focus_index == 2 and self.output_devices:
+            self.output_index = cycle_index(self.output_index, -1, len(self.output_devices))
+            self.speak(self.output_devices[self.output_index]["name"])
+        elif self.focus_index == 3 and self.input_devices:
+            self.input_index = cycle_index(self.input_index, -1, len(self.input_devices))
+            self.speak(self.input_devices[self.input_index]["name"])
+
+    def handle_down(self):
+        if self.focus_index in (0, 1):
+            self.apply_volume_delta(-5)
+            return
+
+        if self.focus_index == 2 and self.output_devices:
+            self.output_index = cycle_index(self.output_index, 1, len(self.output_devices))
+            self.speak(self.output_devices[self.output_index]["name"])
+        elif self.focus_index == 3 and self.input_devices:
+            self.input_index = cycle_index(self.input_index, 1, len(self.input_devices))
+            self.speak(self.input_devices[self.input_index]["name"])
+
+    def start_pending_device_change(self, device_kind, previous_target, candidate):
+        try:
+            self.backend.set_default_device(device_kind, candidate["node_name"])
+            self.pending_change = PendingDeviceChange(
+                device_kind="output device" if device_kind == "sink" else "microphone device",
+                previous_target=previous_target,
+                candidate_target=candidate["node_name"],
+                candidate_name=candidate["name"],
+                started_at=time.time(),
+            )
+
+            if device_kind == "sink":
+                self.default_output_id = candidate["id"]
+                self.default_output_name = candidate["node_name"]
+                self.output_volume = self.backend.get_device_volume("sink", self.default_output_name)
+                self.feedback.play_test_beep()
+            else:
+                self.default_input_id = candidate["id"]
+                self.default_input_name = candidate["node_name"]
+                self.input_volume = self.backend.get_device_volume("source", self.default_input_name)
+
+            self.set_status(
+                f"{self.pending_change.device_kind.capitalize()} changed to {candidate['name']}. Press enter within 15 seconds to keep it.",
+                speak_message=True,
+            )
+        except Exception as error:
+            self.set_status(f"Device change failed: {error}", speak_message=True)
+
+    def confirm_pending_change(self):
+        if self.pending_change is None:
+            return
+
+        message = f"{self.pending_change.device_kind.capitalize()} confirmed."
+        self.pending_change = None
+        self.refresh_state()
+        self.set_status(message, speak_message=True)
+
+    def rollback_pending_change(self):
+        if self.pending_change is None:
+            return
+
+        try:
+            change = self.pending_change
+            rollback_kind = "sink" if change.device_kind == "output device" else "source"
+            self.backend.set_default_device(rollback_kind, change.previous_target)
+            self.pending_change = None
+            self.refresh_state()
+            self.set_status(
+                f"Restored previous {change.device_kind}.",
+                speak_message=True,
+            )
+        except Exception as error:
+            self.pending_change = None
+            self.set_status(f"Failed to restore previous device: {error}", speak_message=True)
+
+    def activate_selected_device(self):
+        if self.pending_change is not None:
+            self.confirm_pending_change()
+            return
+
+        if self.focus_index == 2:
+            candidate = self.get_selected_output_device()
+            if candidate is None:
+                return
+            if candidate["id"] == self.default_output_id:
+                self.speak(f"Already using {candidate['name']}")
+                return
+            self.start_pending_device_change("sink", self.default_output_name, candidate)
+        elif self.focus_index == 3:
+            candidate = self.get_selected_input_device()
+            if candidate is None:
+                return
+            if candidate["id"] == self.default_input_id:
+                self.speak(f"Already using {candidate['name']}")
+                return
+            self.start_pending_device_change("source", self.default_input_name, candidate)
+        else:
+            self.announce_current_focus()
+
+    def draw(self):
+        self.stdscr.clear()
+        self.stdscr.addstr(1, 2, "Stormux Audio Manager", curses.A_BOLD)
+
+        output_device = self.get_selected_output_device()
+        input_device = self.get_selected_input_device()
+        rows = [
+            f"Output Volume: {format_percent_text(self.output_volume)}",
+            f"Microphone Volume: {format_percent_text(self.input_volume)}",
+            f"Output Device: {'' if output_device is None else output_device['name']}",
+            f"Microphone Device: {'' if input_device is None else input_device['name']}",
+        ]
+
+        for row_index, row_text in enumerate(rows):
+            attribute = curses.A_REVERSE if row_index == self.focus_index else curses.A_NORMAL
+            self.stdscr.addstr(4 + row_index, 2, row_text, attribute)
+
+        help_text = "Left/Right: Move  Up/Down: Change  Space: Speak  Enter: Apply or Confirm  H: Help  Q/Esc: Exit"
+        self.stdscr.addstr(10, 2, help_text)
+
+        if self.pending_change is not None:
+            seconds_left = max(
+                0,
+                self.pending_change.timeout_seconds - int(time.time() - self.pending_change.started_at),
+            )
+            self.stdscr.addstr(
+                12,
+                2,
+                f"Pending {self.pending_change.device_kind}: confirm within {seconds_left} seconds.",
+                curses.A_BOLD,
+            )
+
+        if self.status_message:
+            self.stdscr.addstr(14, 2, self.status_message)
+
+        self.stdscr.refresh()
+
+    def cleanup(self):
+        self.feedback.stop_sound()
+        if self.speech_client is not None:
+            try:
+                self.speech_client.close()
+            except Exception:
+                pass
+        if self.stdscr is not None:
+            try:
+                curses.nocbreak()
+                self.stdscr.keypad(False)
+                curses.echo()
+                curses.endwin()
+            except Exception:
+                pass
+
+    def run(self):
+        try:
+            self.init_speech()
+            self.refresh_state()
+
+            self.stdscr = curses.initscr()
+            curses.noecho()
+            curses.cbreak()
+            self.stdscr.keypad(True)
+            self.stdscr.timeout(250)
+            try:
+                curses.curs_set(0)
+            except Exception:
+                pass
+
+            self.speak("Stormux audio manager")
+
+            while True:
+                if pending_change_expired(self.pending_change):
+                    self.rollback_pending_change()
+
+                self.draw()
+                key = self.stdscr.getch()
+
+                if key == -1:
+                    continue
+                if key == curses.KEY_LEFT:
+                    self.move_focus(-1)
+                elif key == curses.KEY_RIGHT:
+                    self.move_focus(1)
+                elif key == curses.KEY_UP:
+                    self.handle_up()
+                elif key == curses.KEY_DOWN:
+                    self.handle_down()
+                elif key in (10, 13, curses.KEY_ENTER):
+                    self.activate_selected_device()
+                elif key == ord(" "):
+                    self.announce_current_focus()
+                elif key in (ord("h"), ord("H")):
+                    self.speak(HELP_TEXT, interrupt=False)
+                elif key in (ord("q"), ord("Q"), 27):
+                    self.rollback_pending_change()
+                    break
+        finally:
+            self.cleanup()
+
+
+def main():
+    try:
+        AudioManagerApp().run()
+    except Exception as error:
+        print(f"Audio manager could not start: {error}", file=sys.stderr)
+        sys.exit(1)
+
+
+if __name__ == "__main__":
+    main()

usr/local/bin/game_launcher.py

@@ -47,6 +47,7 @@ class VoicedMenu:
 
         # Default settings
         self.speechRate = 0  # Normal speech rate (0 is default in speechd)
+        self.speechPitch = 0  # Normal speech pitch
         self.volume = 50      # Default volume level
 
         # Load settings
@@ -97,8 +98,9 @@ class VoicedMenu:
             self.speechClient.set_priority(speechd.Priority.IMPORTANT)
             self.speechClient.set_punctuation(speechd.PunctuationMode.SOME)
 
-            # Apply speech rate from settings
+            # Apply speech settings from saved values
             self.speechClient.set_rate(self.speechRate)
+            self.speechClient.set_pitch(self.speechPitch)
         except Exception as e:
             print(f"Could not initialize speech: {e}")
             # Fallback to None - the speak method will handle this
@@ -116,6 +118,7 @@ class VoicedMenu:
             # Load speech settings
             if 'Speech' in self.config:
                 self.speechRate = self.config.getint('Speech', 'rate', fallback=0)
+                self.speechPitch = self.config.getint('Speech', 'pitch', fallback=0)
 
             # Load volume settings
             if 'Volume' in self.config:
@@ -134,6 +137,7 @@ class VoicedMenu:
             self.config['Speech'] = {}
 
         self.config['Speech']['rate'] = str(self.speechRate)
+        self.config['Speech']['pitch'] = str(self.speechPitch)
 
         # Save volume settings
         if 'Volume' not in self.config:
@@ -174,6 +178,32 @@ class VoicedMenu:
         # Save the new setting
         self.save_settings()
 
+    def increase_speech_pitch(self):
+        """Increase speech pitch"""
+        self.speechPitch = min(100, self.speechPitch + 10)  # Max is 100
+        if self.speechClient:
+            try:
+                self.speechClient.set_pitch(self.speechPitch)
+                self.speak(f"Speech pitch: {self.speechPitch}")
+            except Exception as e:
+                print(f"Error adjusting speech pitch: {e}")
+
+        # Save the new setting
+        self.save_settings()
+
+    def decrease_speech_pitch(self):
+        """Decrease speech pitch"""
+        self.speechPitch = max(-100, self.speechPitch - 10)  # Min is -100
+        if self.speechClient:
+            try:
+                self.speechClient.set_pitch(self.speechPitch)
+                self.speak(f"Speech pitch: {self.speechPitch}")
+            except Exception as e:
+                print(f"Error adjusting speech pitch: {e}")
+
+        # Save the new setting
+        self.save_settings()
+
     def get_current_volume(self):
         """Get the current system volume percentage"""
         try:
@@ -774,6 +804,8 @@ class VoicedMenu:
         B key: Report battery status.
         Left bracket: Decrease speech rate.
         Right bracket: Increase speech rate.
+        Left brace: Decrease speech pitch.
+        Right brace: Increase speech pitch.
         9 key: Decrease volume.
         0 key: Increase volume.
         Escape: Refresh the menu.
@@ -793,7 +825,7 @@ class VoicedMenu:
         self.stdscr.addstr(1, x, title, curses.A_BOLD)
 
         # Draw help line
-        helpText = "Ãvigate | Enter: Select | H: Help | [ ] Rate | 9 0 Volume | Esc: Refresh"
+        helpText = "Navigate | Enter: Select | H: Help | [ ] Rate | { } Pitch | 9 0 Volume | Esc: Refresh"
         x = max(0, w // 2 - len(helpText) // 2)
         self.stdscr.addstr(3, x, helpText)
 
@@ -826,6 +858,11 @@ class VoicedMenu:
         rateText = f"Speech Rate: {self.speechRate}"
         self.stdscr.addstr(h-2, 2, rateText)
 
+        # Draw speech pitch indicator
+        pitchText = f"Pitch: {self.speechPitch}"
+        pitchX = max(0, w // 2 - len(pitchText) // 2)
+        self.stdscr.addstr(h-2, pitchX, pitchText)
+
         # Draw volume indicator
         volumeText = f"Volume: {self.get_current_volume()}%"
         self.stdscr.addstr(h-2, w-len(volumeText)-2, volumeText)
@@ -965,6 +1002,14 @@ class VoicedMenu:
                     self.increase_speech_rate()
                     self.draw_menu()
 
+                elif key == ord('{'):  # Decrease speech pitch
+                    self.decrease_speech_pitch()
+                    self.draw_menu()
+
+                elif key == ord('}'):  # Increase speech pitch
+                    self.increase_speech_pitch()
+                    self.draw_menu()
+
                 elif key == ord('9'):  # Decrease volume
                     self.decrease_volume()
 
@@ -1147,6 +1192,7 @@ if __name__ == "__main__":
     menu.add_section("Accessories")
     menu.add_item("Accessories", "Local IP Address", "/usr/local/bin/ip_info.py local")
     menu.add_item("Accessories", "Remote IP Address", "/usr/local/bin/ip_info.py remote")
+    menu.add_item("Accessories", "Sound and Volume", "/usr/local/bin/audio_manager.py")
     menu.add_item("Accessories", "Web Browser", "GAME=Brave startx")
     menu.add_item("Accessories", "LibreOffice", lambda: menu.install_and_launch("libreoffice", "gui"))
     menu.add_item("Accessories", "Thunderbird", lambda: menu.install_and_launch("thunderbird", "gui"))