Add configurable incoming audio jitter buffer
This commit is contained in:
committed by
Brandon McGinty
parent
d338925da6
commit
c0ef036934
@@ -154,6 +154,21 @@ barnard --audio-interval 20
|
||||
Supported values are `10`, `20`, `40`, and `60` milliseconds. Try `20` ms
|
||||
first; use `40` ms only if the connection remains unreliable.
|
||||
|
||||
## Incoming Audio Jitter Buffer
|
||||
|
||||
Barnard holds 40 ms of audio separately for each speaker before starting
|
||||
playback. This prevents brief delayed UDP packets from draining OpenAL's audio
|
||||
queue, which otherwise produces clicks or pops. To adjust this tradeoff between
|
||||
resilience and added incoming latency:
|
||||
|
||||
```sh
|
||||
barnard --jitter-buffer 60
|
||||
```
|
||||
|
||||
Supported values are `0`, `20`, `40` (default), and `60` milliseconds. Try
|
||||
`60` ms for a lossy or jittery connection. Use `0` only when minimizing latency
|
||||
is more important than avoiding playback underruns.
|
||||
|
||||
## Audio Devices
|
||||
|
||||
You can set the default input and output devices in the config file as well.
|
||||
|
||||
@@ -86,6 +86,22 @@ func TestAudioIntervalDuration(t *testing.T) {
|
||||
}
|
||||
}
|
||||
|
||||
func TestJitterBufferDuration(t *testing.T) {
|
||||
for _, milliseconds := range []int{0, 20, 40, 60} {
|
||||
got, err := jitterBufferDuration(milliseconds)
|
||||
if err != nil {
|
||||
t.Errorf("jitterBufferDuration(%d): %v", milliseconds, err)
|
||||
continue
|
||||
}
|
||||
if got != time.Duration(milliseconds)*time.Millisecond {
|
||||
t.Errorf("jitterBufferDuration(%d) = %v", milliseconds, got)
|
||||
}
|
||||
}
|
||||
if _, err := jitterBufferDuration(10); err == nil {
|
||||
t.Fatal("jitterBufferDuration accepted unsupported duration")
|
||||
}
|
||||
}
|
||||
|
||||
func TestServerAddressDefaultsPortWithoutBreakingIPv6(t *testing.T) {
|
||||
for input, want := range map[string]string{
|
||||
"server": "server:64738",
|
||||
|
||||
+10
-3
@@ -25,6 +25,9 @@ type Config struct {
|
||||
AudioInterval time.Duration
|
||||
// AudioDataBytes is the number of bytes that an audio frame can use.
|
||||
AudioDataBytes int
|
||||
// IncomingAudioBuffer is the amount of per-speaker audio retained before
|
||||
// playback starts, absorbing jitter in incoming UDP packet delivery.
|
||||
IncomingAudioBuffer time.Duration
|
||||
|
||||
// DisableUDP forces all audio to use the TCP tunnel instead of UDP.
|
||||
DisableUDP bool
|
||||
@@ -38,9 +41,10 @@ type Config struct {
|
||||
// NewConfig returns a new Config struct with default values set.
|
||||
func NewConfig() *Config {
|
||||
return &Config{
|
||||
Buffers: 8,
|
||||
AudioInterval: AudioDefaultInterval,
|
||||
AudioDataBytes: AudioDefaultDataBytes,
|
||||
Buffers: 8,
|
||||
AudioInterval: AudioDefaultInterval,
|
||||
AudioDataBytes: AudioDefaultDataBytes,
|
||||
IncomingAudioBuffer: 40 * time.Millisecond,
|
||||
}
|
||||
}
|
||||
|
||||
@@ -54,6 +58,9 @@ func (c *Config) Validate() error {
|
||||
if c.AudioDataBytes <= 0 {
|
||||
return fmt.Errorf("gumble: AudioDataBytes must be positive")
|
||||
}
|
||||
if c.IncomingAudioBuffer < 0 {
|
||||
return fmt.Errorf("gumble: IncomingAudioBuffer must not be negative")
|
||||
}
|
||||
if c.Buffers <= 0 {
|
||||
return fmt.Errorf("gumble: Buffers must be positive")
|
||||
}
|
||||
|
||||
@@ -210,7 +210,6 @@ func (c *Client) handleUDPTunnel(buffer []byte) error {
|
||||
},
|
||||
Sequence: seq,
|
||||
AudioBuffer: AudioBuffer(pcm),
|
||||
Terminator: isFinal,
|
||||
}
|
||||
|
||||
if len(buffer)-audioLength == 3*4 {
|
||||
@@ -227,6 +226,7 @@ func (c *Client) handleUDPTunnel(buffer []byte) error {
|
||||
if isFinal {
|
||||
decoder.Reset()
|
||||
user.audioSequenceValid = false
|
||||
c.dispatchAudio(user, &AudioPacket{Client: c, Sender: user, Terminator: true})
|
||||
}
|
||||
return nil
|
||||
}
|
||||
|
||||
@@ -861,7 +861,6 @@ func (c *Client) decodeAndDispatch(pktNum uint64, user *User, decoder AudioDecod
|
||||
Target: &VoiceTarget{ID: context},
|
||||
Sequence: frameNum,
|
||||
AudioBuffer: AudioBuffer(pcm),
|
||||
Terminator: terminator,
|
||||
VolumeAdjustment: volumeAdjustment,
|
||||
}
|
||||
if position != nil {
|
||||
@@ -873,6 +872,7 @@ func (c *Client) decodeAndDispatch(pktNum uint64, user *User, decoder AudioDecod
|
||||
decoder.Reset()
|
||||
user.audioSequenceValid = false
|
||||
user.audioFrameStep = 0
|
||||
c.dispatchAudio(user, &AudioPacket{Client: c, Sender: user, Terminator: true})
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
@@ -52,14 +52,22 @@ const recorderOutgoingSource uint32 = ^uint32(0)
|
||||
|
||||
const (
|
||||
maxBufferSize = 11520 // Max frame size (2880) * bytes per stereo sample (4)
|
||||
jitterMinPackets = 3
|
||||
jitterMaxPackets = 10
|
||||
jitterMaxPackets = 50
|
||||
)
|
||||
|
||||
// jitterPlaybackReady holds the initial playout delay only once. Requiring
|
||||
// the minimum on every packet drains and refills the renderer in bursts.
|
||||
func jitterPlaybackReady(started bool, buffered int) bool {
|
||||
return started || buffered >= jitterMinPackets
|
||||
// jitterPlaybackReady holds the requested initial playout delay only once.
|
||||
// Requiring the delay on every packet drains and refills the renderer in bursts.
|
||||
func jitterPlaybackReady(started bool, buffered, target time.Duration) bool {
|
||||
return started || buffered >= target
|
||||
}
|
||||
|
||||
func audioPacketDuration(packet *gumble.AudioPacket) time.Duration {
|
||||
if packet == nil || len(packet.AudioBuffer) == 0 {
|
||||
return 0
|
||||
}
|
||||
// Opus decoders deliver interleaved stereo PCM to this renderer.
|
||||
frames := len(packet.AudioBuffer) / gumble.AudioChannels
|
||||
return time.Duration(frames) * time.Second / gumble.AudioSampleRate
|
||||
}
|
||||
|
||||
var (
|
||||
@@ -484,11 +492,13 @@ func (s *Stream) OnAudioStream(e *gumble.AudioStreamEvent) {
|
||||
// Jitter buffer: collects incoming packets, reorders by
|
||||
// sequence number, and releases them after a small initial delay.
|
||||
var jitterBuf []*gumble.AudioPacket
|
||||
var jitterDuration time.Duration
|
||||
var jitterNextSeq int64
|
||||
var jitterInit, jitterStarted bool
|
||||
var jitterDrainLogCounter, jitterAnomalyLogCounter int
|
||||
resetJitter := func() {
|
||||
jitterBuf = nil
|
||||
jitterDuration = 0
|
||||
jitterNextSeq = 0
|
||||
jitterInit = false
|
||||
jitterStarted = false
|
||||
@@ -513,6 +523,7 @@ func (s *Stream) OnAudioStream(e *gumble.AudioStreamEvent) {
|
||||
jitterBuf = append(jitterBuf, nil)
|
||||
copy(jitterBuf[i+1:], jitterBuf[i:])
|
||||
jitterBuf[i] = p
|
||||
jitterDuration += audioPacketDuration(p)
|
||||
}
|
||||
|
||||
// popNext removes and returns the packet with the expected next
|
||||
@@ -523,6 +534,7 @@ func (s *Stream) OnAudioStream(e *gumble.AudioStreamEvent) {
|
||||
}
|
||||
p := jitterBuf[0]
|
||||
jitterBuf = jitterBuf[1:]
|
||||
jitterDuration -= audioPacketDuration(p)
|
||||
// Frame numbers are Mumble timestamps in 10 ms units.
|
||||
// Compute the actual step from the PCM sample count so we
|
||||
// never skip a legitimate gap.
|
||||
@@ -571,8 +583,8 @@ func (s *Stream) OnAudioStream(e *gumble.AudioStreamEvent) {
|
||||
|
||||
// Hold only the initial packets. Once playback starts, drain every
|
||||
// ready packet so the renderer is fed continuously rather than in
|
||||
// bursts of jitterMinPackets packets.
|
||||
if !jitterPlaybackReady(jitterStarted, len(jitterBuf)) {
|
||||
// bursts of packets.
|
||||
if !jitterPlaybackReady(jitterStarted, jitterDuration, e.Client.Config.IncomingAudioBuffer) {
|
||||
continue
|
||||
}
|
||||
jitterStarted = true
|
||||
@@ -590,6 +602,7 @@ func (s *Stream) OnAudioStream(e *gumble.AudioStreamEvent) {
|
||||
log.Debug("jitter: discarding late seq=%d for %s (next=%d buf=%d)",
|
||||
jitterBuf[0].Sequence, e.User.Name, jitterNextSeq, len(jitterBuf))
|
||||
}
|
||||
jitterDuration -= audioPacketDuration(jitterBuf[0])
|
||||
jitterBuf = jitterBuf[1:]
|
||||
continue
|
||||
}
|
||||
|
||||
@@ -4,8 +4,10 @@ import (
|
||||
"errors"
|
||||
"strings"
|
||||
"testing"
|
||||
"time"
|
||||
|
||||
"git.stormux.org/storm/barnard/gumble/go-openal/openal"
|
||||
"git.stormux.org/storm/barnard/gumble/gumble"
|
||||
)
|
||||
|
||||
// Regression: audio cleanup could send a final render command after Destroy
|
||||
@@ -49,17 +51,24 @@ func TestStopSourceWaitsForWorker(t *testing.T) {
|
||||
}
|
||||
|
||||
func TestJitterPlaybackDelayAppliesOnlyAtStartup(t *testing.T) {
|
||||
if jitterPlaybackReady(false, jitterMinPackets-1) {
|
||||
if jitterPlaybackReady(false, 20*time.Millisecond, 40*time.Millisecond) {
|
||||
t.Fatal("jitter playback started before initial buffer filled")
|
||||
}
|
||||
if !jitterPlaybackReady(false, jitterMinPackets) {
|
||||
if !jitterPlaybackReady(false, 40*time.Millisecond, 40*time.Millisecond) {
|
||||
t.Fatal("jitter playback did not start after initial buffer filled")
|
||||
}
|
||||
if !jitterPlaybackReady(true, 1) {
|
||||
if !jitterPlaybackReady(true, 0, 40*time.Millisecond) {
|
||||
t.Fatal("jitter playback paused while refilling after startup")
|
||||
}
|
||||
}
|
||||
|
||||
func TestAudioPacketDurationUsesStereoFrameCount(t *testing.T) {
|
||||
packet := &gumble.AudioPacket{AudioBuffer: make(gumble.AudioBuffer, 2*gumble.AudioDefaultFrameSize)}
|
||||
if got := audioPacketDuration(packet); got != 10*time.Millisecond {
|
||||
t.Fatalf("audioPacketDuration = %v, want 10ms", got)
|
||||
}
|
||||
}
|
||||
|
||||
func TestRenderRejectsWorkAfterShutdown(t *testing.T) {
|
||||
s := &Stream{renderClosed: true}
|
||||
called := false
|
||||
|
||||
@@ -86,6 +86,7 @@ func main() {
|
||||
certificateSet := false
|
||||
buffers := flag.Int("buffers", 16, "number of audio buffers to use")
|
||||
audioInterval := flag.Int("audio-interval", 10, "outgoing audio packet duration in ms (10, 20, 40, or 60)")
|
||||
jitterBuffer := flag.Int("jitter-buffer", 40, "incoming per-user audio buffer in ms (0, 20, 40, or 60)")
|
||||
profile := flag.Bool("profile", false, "add http server to serve profiles")
|
||||
noiseSuppressionEnabled := flag.Bool("noise-suppression", false, "enable noise suppression for microphone input")
|
||||
autoTransmit := flag.Bool("auto-transmit", false, "start transmitting immediately on connect")
|
||||
@@ -100,6 +101,10 @@ func main() {
|
||||
if err != nil {
|
||||
handle_raw_error(err)
|
||||
}
|
||||
selectedJitterBuffer, err := jitterBufferDuration(*jitterBuffer)
|
||||
if err != nil {
|
||||
handle_raw_error(err)
|
||||
}
|
||||
|
||||
// Set up logging
|
||||
var level barnlog.Level
|
||||
@@ -199,6 +204,7 @@ func main() {
|
||||
}
|
||||
b.Config.Buffers = *buffers
|
||||
b.Config.AudioInterval = selectedAudioInterval
|
||||
b.Config.IncomingAudioBuffer = selectedJitterBuffer
|
||||
b.Config.DisableUDP = *tcpOnly
|
||||
|
||||
b.Hotkeys = b.UserConfig.GetHotkeys()
|
||||
@@ -257,6 +263,18 @@ func audioIntervalDuration(milliseconds int) (time.Duration, error) {
|
||||
}
|
||||
}
|
||||
|
||||
// jitterBufferDuration converts the requested incoming playout delay to a
|
||||
// supported duration. Zero starts playback without an initial safety buffer.
|
||||
func jitterBufferDuration(milliseconds int) (time.Duration, error) {
|
||||
interval := time.Duration(milliseconds) * time.Millisecond
|
||||
switch interval {
|
||||
case 0, 20 * time.Millisecond, 40 * time.Millisecond, 60 * time.Millisecond:
|
||||
return interval, nil
|
||||
default:
|
||||
return 0, fmt.Errorf("jitter buffer must be 0, 20, 40, or 60 ms, got %d", milliseconds)
|
||||
}
|
||||
}
|
||||
|
||||
// serverAddress adds Mumble's default port without corrupting an IPv6 literal.
|
||||
func serverAddress(address string) string {
|
||||
if _, port, err := net.SplitHostPort(address); err == nil && port != "" {
|
||||
|
||||
Reference in New Issue
Block a user