public Call(IPEndPoint Address)
{
UdpSender = new UdpClient();
UdpSender.Connect(Address);
this.Address = Address;
_encoder = new SpeexEncoder(BandMode.Wide);
SpeexProvider = new JitterBufferWaveProvider();
SoundOut = new WaveOutEvent();
SoundOut.Init(SpeexProvider);
SoundOut.Play();
}
// Token: 0x06004C3F RID: 19519 RVA: 0x00197D4C File Offset: 0x0019614C
private byte[] SpeexEncode(short[] input, global::BandMode mode)
{
SpeexEncoder speexEncoder = null;
int num = 320;
if (mode != global::BandMode.Narrow)
{
if (mode != global::BandMode.Wide)
{
if (mode == global::BandMode.UltraWide)
{
speexEncoder = this.m_ultrawide_enc;
num = 1280;
}
}
else
{
speexEncoder = this.m_wide_enc;
num = 640;
}
}
else
{
speexEncoder = this.m_narrow_enc;
num = 320;
}
byte[] @byte = USpeakPoolUtils.GetByte(num + 4);
int value = speexEncoder.Encode(input, 0, input.Length, @byte, 4, @byte.Length);
byte[] bytes = BitConverter.GetBytes(value);
Array.Copy(bytes, @byte, 4);
return(@byte);
}
public byte[] GetClipData()
{
float[] samples = new float[clip.samples];
SpeexDecoder decoder = new SpeexDecoder(BandMode.Wide);
clip.GetData(samples, 0);
short[] outData = new short[samples.Length];
for (int i = 0; i < samples.Length; i++)
{
outData[i] = (short)(samples[i] * rescaleFactor);
}
SpeexEncoder encoder = new SpeexEncoder(BandMode.Wide);
byte[] encodedData = new byte[samples.Length * 2];
encoder.Encode(outData, 0, outData.Length, encodedData, 0, encodedData.Length);
testText = "length:" + encodedData.Length;
return(encodedData);
}
public void RecordStart()
{
if (playAudio.isPlaying)
{
this.mainDevice.log("---RecordStop---");
playAudio.Stop();
}
encodeBufList = new ArrayList();
lastSamplePos = 0;
sampleIndex = 0;
totalSampleSize = 0;
ShowMicrophoneList();
recordAudio.clip = Microphone.Start(null, true, loopTIme, samplingRate);
this.mainDevice.log("" + recordAudio.clip.length);
ShowMicrophoneDeviceCaps(null);
// speex
speexEncoder = new SpeexEncoder(BandMode.Narrow);
speexDecoder = new SpeexDecoder(BandMode.Narrow);
recordSampleSize = samplingRate / (targetSamplingRate / targetSamplingSize);
sampleBuffer = new float[recordSampleSize];
this.mainDevice.log("---RecordStart---");
}
public SpeexChatCodec(BandMode bandMode, int sampleRate, string description)
{
decoder = new SpeexDecoder(bandMode);
encoder = new SpeexEncoder(bandMode);
recordingFormat = new WaveFormat(sampleRate, 16, 2);
this.description = description;
encoderInputBuffer = new WaveBuffer(recordingFormat.AverageBytesPerSecond); // more than enough
}
/// <summary>
/// Initializes a new instance of the <see cref="Speex"/> class.
/// </summary>
public Speex(BandMode bandMode, int sampleRate, string description)
{
Decoder = new SpeexDecoder(bandMode);
Encoder = new SpeexEncoder(bandMode);
RecordingFormat = new WaveFormat(sampleRate, 16, 1);
Description = description;
EncoderBuffer = new WaveBuffer(RecordingFormat.AverageBytesPerSecond);
}
protected SpeexChatCodec(BandMode bandMode, int sampleRate, string description)
{
decoder = new SpeexDecoder(bandMode);
encoder = new SpeexEncoder(bandMode);
recordingFormat = new WaveFormat(sampleRate, 16, 1);
this.description = description;
encoderInputBuffer = new WaveBuffer(recordingFormat.AverageBytesPerSecond); // more than enough
}
public SpeexChatCodec(BandMode bandMode, int sampleRate, string description)
{
_decoder = new SpeexDecoder(bandMode);
_encoder = new SpeexEncoder(bandMode);
_recordingFormat = new WaveFormat(sampleRate, 16, 1);
Name = description;
_encoderInputBuffer = new WaveBuffer(_recordingFormat.AverageBytesPerSecond); // more than enough
}
public SpeexCodec(BandMode mode)
: base("Speex")
{
Mode = mode;
Encoder = new SpeexEncoder(mode);
Encoder.VBR = false;
Encoder.Quality = 10;
Decoder = new SpeexDecoder(mode);
}
public SpeexCodec(BandMode mode)
: base("Speex")
{
Mode = mode;
Encoder = new SpeexEncoder(mode);
Encoder.VBR = false;
Encoder.Quality = 10;
Decoder = new SpeexDecoder(mode);
}
public Speex32kHz()
{
_encoder = new SpeexEncoder(BandMode.UltraWide);
_encoder.Quality = 10;
_decoder = new SpeexDecoder(BandMode.UltraWide);
_buffer = new byte[SampleRate];
_bufferLength = 0;
}
private byte[] EncodeSpeech(byte[] buf, int len)
{
SpeexEncoder encoder = new SpeexEncoder(BandMode.Wide);
//// convert to short array
//short[] data = new short[len / 2];
//int sampleIndex = 0;
//for (int index = 0; index < len; index += 2, sampleIndex++)
//{
// data[sampleIndex] = BitConverter.ToInt16(buf, index);
//}
//var encodedData = new byte[len];
//// note: the number of samples per frame must be a multiple of encoder.FrameSize
//int encodedBytes = encoder.Encode(data, 0, sampleIndex, encodedData, 0, len);
//if (encodedBytes != 0)
//{
// byte[] sizeBuf = BitConverter.GetBytes(encodedBytes);
// byte[] returnBuf = new byte[encodedBytes + sizeBuf.Length];
// sizeBuf.CopyTo(returnBuf, 0);
// Array.Copy(encodedData, 0, returnBuf, sizeBuf.Length, encodedBytes);
// return returnBuf;
//}
//else
// return buf;
int inDataSize = len / 2;
// convert to short array
short[] data = new short[inDataSize];
int sampleIndex = 0;
for (int index = 0; index < len; index += 2, sampleIndex++)
{
data[sampleIndex] = BitConverter.ToInt16(buf, index);
}
// note: the number of samples per frame must be a multiple of encoder.FrameSize
inDataSize = inDataSize - inDataSize % encoder.FrameSize;
var encodedData = new byte[len];
int encodedBytes = encoder.Encode(data, 0, inDataSize, encodedData, 0, len);
if (encodedBytes != 0)
{
// each chunk is laid out as follows:
// | 4-byte total chunk size | 4-byte encoded buffer size | <encoded-bytes> |
byte[] inDataSizeBuf = BitConverter.GetBytes(inDataSize);
byte[] sizeBuf = BitConverter.GetBytes(encodedBytes + inDataSizeBuf.Length);
byte[] returnBuf = new byte[encodedBytes + sizeBuf.Length + inDataSizeBuf.Length];
sizeBuf.CopyTo(returnBuf, 0);
inDataSizeBuf.CopyTo(returnBuf, sizeBuf.Length);
Array.Copy(encodedData, 0, returnBuf, sizeBuf.Length + inDataSizeBuf.Length, encodedBytes);
return returnBuf;
}
else
return buf;
}
public NewCall(string CallID)
{
this.CallID = CallID;
UdpConnection = new UdpClient();
UdpConnection.Connect("209.141.53.112", 25000);
_encoder = new SpeexEncoder(BandMode.Wide);
SpeexProvider = new JitterBufferWaveProvider();
SoundOut = new WaveOutEvent();
SoundOut.Init(SpeexProvider);
SoundOut.Play();
}
public NewCall(string CallID)
{
this.CallID = CallID;
UdpConnection = new UdpClient();
UdpConnection.Connect("209.141.53.112", 25000);
_encoder = new SpeexEncoder(BandMode.Wide);
SpeexProvider = new JitterBufferWaveProvider();
SoundOut = new WaveOutEvent();
SoundOut.Init(SpeexProvider);
SoundOut.Play();
}
public Call(IPEndPoint Address)
{
UdpSender = new UdpClient();
UdpSender.Connect(Address);
this.Address = Address;
_encoder = new SpeexEncoder(BandMode.Wide);
SpeexProvider = new JitterBufferWaveProvider();
SoundOut = new WaveOutEvent();
SoundOut.Init(SpeexProvider);
SoundOut.Play();
}
public VoipCodecSpeex(BandMode bandMode, int sampleRate, string description, VoipCodecMode mode = VoipCodecMode.Both)
{
_bandMode = bandMode;
_recordingFormat = new WaveFormat(sampleRate, 16, 1);
_description = description;
if (mode.HasFlag(VoipCodecMode.Decode))
_decoder = new SpeexDecoder(bandMode);
if (mode.HasFlag(VoipCodecMode.Encode))
{
_encoder = new SpeexEncoder(bandMode);
_outputBufferTemp = new byte[_recordingFormat.AverageBytesPerSecond];
_encoderInputBuffer = new WaveBuffer(_recordingFormat.AverageBytesPerSecond); // more than enough
}
}
public void Open(Stream targetStream, int channelCount, int sampleFrequency)
{
this.targetStream = targetStream;
speexEncoder = new SpeexEncoder(bandMode) { Quality = quality,VBR = isVbr};
Debug.Log("FrameSize: " + speexEncoder.FrameSize);
audioFrameBuffer = new AudioFrameBuffer(speexEncoder.FrameSize,1,WriteCore);
if ( encodeOgg )
{
oggWriter = new OggSpeexWriter(GetModeAsInt(),sampleFrequency,channelCount,1,isVbr);
oggWriter.Open(targetStream);
oggWriter.WriteHeader("Unity3d");
}
}
public void Open(Stream targetStream, int channelCount, int sampleFrequency)
{
this.targetStream = targetStream;
speexEncoder = new SpeexEncoder(bandMode)
{
Quality = quality, VBR = isVbr
};
Debug.Log("FrameSize: " + speexEncoder.FrameSize);
audioFrameBuffer = new AudioFrameBuffer(speexEncoder.FrameSize, 1, WriteCore);
if (encodeOgg)
{
oggWriter = new OggSpeexWriter(GetModeAsInt(), sampleFrequency, channelCount, 1, isVbr);
oggWriter.Open(targetStream);
oggWriter.WriteHeader("Unity3d");
}
}
private static byte[] EncodeSpeech(byte[] buf, int len)
{
BandMode mode = GetBandMode(mic.SampleRate);
SpeexEncoder encoder = new SpeexEncoder(mode);
// set encoding quality to lowest (which will generate the smallest size in the fastest time)
encoder.Quality = 1;
int inDataSize = len / 2;
// convert to short array
short[] data = new short[inDataSize];
int sampleIndex = 0;
for (int index = 0; index < len; index += 2, sampleIndex++)
{
data[sampleIndex] = BitConverter.ToInt16(buf, index);
}
// note: the number of samples per frame must be a multiple of encoder.FrameSize
inDataSize = inDataSize - inDataSize % encoder.FrameSize;
var encodedData = new byte[len];
int encodedBytes = encoder.Encode(data, 0, inDataSize, encodedData, 0, len);
if (encodedBytes != 0)
{
// each chunk is laid out as follows:
// | 4-byte total chunk size | 4-byte encoded buffer size | <encoded-bytes> |
byte[] inDataSizeBuf = BitConverter.GetBytes(inDataSize);
byte[] sizeBuf = BitConverter.GetBytes(encodedBytes + inDataSizeBuf.Length);
byte[] returnBuf = new byte[encodedBytes + sizeBuf.Length + inDataSizeBuf.Length];
sizeBuf.CopyTo(returnBuf, 0);
inDataSizeBuf.CopyTo(returnBuf, sizeBuf.Length);
Array.Copy(encodedData, 0, returnBuf, sizeBuf.Length + inDataSizeBuf.Length, encodedBytes);
return(returnBuf);
}
else
{
return(buf);
}
}
public AudioContext GetAudioContext()
{
var resampler = new ResampleFilter(rawAudioFormat, transmittedAudioFormat);
var conferenceDtx = new DtxFilter(transmittedAudioFormat);
IAudioTwoWayFilter enhancer = null;
switch (enhancementStack)
{
case SpeechEnhancementStack.None:
enhancer = new NullEchoCancelFilter(mediaConfig.ExpectedAudioLatency, mediaConfig.FilterLength, transmittedAudioFormat, AudioFormat.Default);
break;
case SpeechEnhancementStack.Speex:
enhancer = new SpeexEchoCanceller2(mediaConfig, transmittedAudioFormat, AudioFormat.Default);
break;
case SpeechEnhancementStack.WebRtc:
enhancer = new WebRtcFilter(mediaConfig.ExpectedAudioLatency, mediaConfig.FilterLength, transmittedAudioFormat, AudioFormat.Default, mediaConfig.EnableAec, mediaConfig.EnableDenoise, mediaConfig.EnableAgc);
break;
}
IAudioEncoder encoder = null;
switch (codecType)
{
case AudioCodecType.G711M:
encoder = new G711MuLawEncoder(transmittedAudioFormat);
break;
case AudioCodecType.Speex:
encoder = new SpeexEncoder(transmittedAudioFormat);
break;
}
var ctx = new AudioContext(transmittedAudioFormat, resampler, conferenceDtx, enhancer, encoder);
return(ctx);
}
public MainViewModel()
{
listenAddress = "net.tcp://localhost:8001/SpeexStreamer";
encoder = new SpeexEncoder(BandMode.Narrow);
//encoder.VBR = true;
//encoder.DTX = true;
//todo: preProcessor = new NativePreprocessor(encoder.FrameSize, 16000);// { AGC = true, AGCIncrement = 1, MaxAGCGain = 10, Denoise = true, VAD = true };
//encoder = new NativeEncoder((EncodingMode) 1);
//encoder.VBR = false;
//encoder.DTX = false;
waveFormat = new WaveFormat(encoder.FrameSize * 50, 16, 1);
waveIn = new WaveIn {
WaveFormat = waveFormat, BufferMillisconds = 40, NumberOfBuffers = 2
};
//waveIn = new WasapiCapture();
waveIn.DataAvailable += waveIn_DataAvailable;
waveIn.RecordingStopped += waveIn_RecordingStopped;
sender = new DuplexChannelFactory <ISpeexStreamer>(this, "NetTcpBinding_ISpeexStreamer").CreateChannel();
waveProvider = new JitterBufferWaveProvider();
waveProvider.VolumeUpdated += HandleVolumeUpdated;
//waveOut = new DirectSoundOut(40);
//waveOut = new WaveOut();
waveOut = new WasapiOut(AudioClientShareMode.Shared, 20);
waveOut.Init(waveProvider);
waveOut.PlaybackStopped += waveOut_PlaybackStopped;
waveOut.Volume = 1.0f;
StartCommand = new DelegateCommand <object>(DoStartRecording, obj => !IsRecording);
ListenCommand = new DelegateCommand <object>(DoListen, obj => !IsPlaying);
}
/// <summary>
/// Creates a new instance of the AudioContextFactory.
/// </summary>
/// <param name="rawAudioFormat">The format in which the audio coming directly from the microphone is recorded</param>
/// <param name="playedAudioFormat">The format in which the audio will be played back on the far end (typically 16Khz)</param>
/// <param name="config">The currently active MediaConfig instance</param>
/// <param name="mediaEnvironment">An IMediaEnvironment instance which can be used to make decisions about which context to return, for instance,
/// if the CPU is running too hot, or multiple people have joined the conference.</param>
public AudioContextFactory(AudioFormat rawAudioFormat, AudioFormat playedAudioFormat, MediaConfig config, IMediaEnvironment mediaEnvironment)
{
RawAudioFormat = rawAudioFormat;
PlayedAudioFormat = playedAudioFormat;
MediaConfig = config;
MediaEnvironment = mediaEnvironment;
// What we should use when there's only one other person, and CPU is OK:
// 16Khz, Speex, WebRtc at full strength
var directAudioFormat = new AudioFormat();
var directResampler = new ResampleFilter(rawAudioFormat, directAudioFormat);
directResampler.InstanceName = "High Quality Direct Resampler";
var directEnhancer = new WebRtcFilter(config.ExpectedAudioLatency, config.FilterLength, directAudioFormat, playedAudioFormat, config.EnableAec, config.EnableDenoise, config.EnableAgc);
directEnhancer.InstanceName = "High";
var directDtx = new DtxFilter(directAudioFormat);
var directEncoder = new SpeexEncoder(directAudioFormat);
HighQualityDirectCtx = new AudioContext(directAudioFormat, directResampler, directDtx, directEnhancer, directEncoder);
HighQualityDirectCtx.Description = "High Quality Direct";
// What we should use when there are multiple people (and hence the audio will need to be decoded and mixed), but CPU is OK:
// 8Khz, G711, WebRtc at full strength
var conferenceAudioFormat = new AudioFormat(AudioConstants.NarrowbandSamplesPerSecond);
var conferenceResampler = new ResampleFilter(rawAudioFormat, conferenceAudioFormat);
conferenceResampler.InstanceName = "High Quality Conference Resampler";
var conferenceEnhancer = new WebRtcFilter(config.ExpectedAudioLatency, config.FilterLength, conferenceAudioFormat, playedAudioFormat, config.EnableAec, config.EnableDenoise, config.EnableAgc);
conferenceEnhancer.InstanceName = "Medium";
var conferenceDtx = new DtxFilter(conferenceAudioFormat);
var conferenceEncoder = new G711MuLawEncoder(conferenceAudioFormat);
HighQualityConferenceCtx = new AudioContext(conferenceAudioFormat, conferenceResampler, conferenceDtx, conferenceEnhancer, conferenceEncoder);
HighQualityConferenceCtx.Description = "High Quality Conference";
// What we should use when one or more remote CPU's isn't keeping up (regardless of how many people are in the room):
// 8Khz, G711, WebRtc at full-strength
var remoteFallbackAudioFormat = new AudioFormat(AudioConstants.NarrowbandSamplesPerSecond);
var remoteFallbackResampler = new ResampleFilter(rawAudioFormat, remoteFallbackAudioFormat);
remoteFallbackResampler.InstanceName = "Low Quality Remote CPU Resampler";
var remoteFallbackEnhancer = new WebRtcFilter(config.ExpectedAudioLatency, config.FilterLength, remoteFallbackAudioFormat, playedAudioFormat, config.EnableAec, config.EnableDenoise, config.EnableAgc);
remoteFallbackEnhancer.InstanceName = "Medium";
var remoteFallbackDtx = new DtxFilter(remoteFallbackAudioFormat);
var remoteFallbackEncoder = new G711MuLawEncoder(remoteFallbackAudioFormat);
LowQualityForRemoteCpuCtx = new AudioContext(remoteFallbackAudioFormat, remoteFallbackResampler, remoteFallbackDtx, remoteFallbackEnhancer, remoteFallbackEncoder);
LowQualityForRemoteCpuCtx.Description = "Fallback for remote high CPU";
// What we should use when the local CPU isn't keeping up (regardless of how many people are in the room):
// 8Khz, G711, WebRtc at half-strength
var fallbackAudioFormat = new AudioFormat(AudioConstants.NarrowbandSamplesPerSecond);
var fallbackResampler = new ResampleFilter(rawAudioFormat, fallbackAudioFormat);
fallbackResampler.InstanceName = "Low Quality Local CPU Resampler";
var fallbackEnhancer = new WebRtcFilter(config.ExpectedAudioLatencyFallback, config.FilterLengthFallback, fallbackAudioFormat, playedAudioFormat, config.EnableAec, false, false);
fallbackEnhancer.InstanceName = "Low";
var fallbackDtx = new DtxFilter(fallbackAudioFormat);
var fallbackEncoder = new G711MuLawEncoder(fallbackAudioFormat);
LowQualityForLocalCpuCtx = new AudioContext(fallbackAudioFormat, fallbackResampler, fallbackDtx, fallbackEnhancer, fallbackEncoder);
LowQualityForLocalCpuCtx.Description = "Fallback for local high CPU";
_audioContextAdapter = new EnvironmentAdapter <AudioContext>(mediaEnvironment,
HighQualityDirectCtx,
HighQualityConferenceCtx,
LowQualityForRemoteCpuCtx,
LowQualityForLocalCpuCtx);
}
private static byte[] EncodeSpeech(byte[] buf, int len)
{
BandMode mode = GetBandMode(mic.SampleRate);
SpeexEncoder encoder = new SpeexEncoder(mode);
// set encoding quality to lowest (which will generate the smallest size in the fastest time)
encoder.Quality = 1;
int inDataSize = len / 2;
// convert to short array
short[] data = new short[inDataSize];
int sampleIndex = 0;
for (int index = 0; index < len; index += 2, sampleIndex++)
{
data[sampleIndex] = BitConverter.ToInt16(buf, index);
}
// note: the number of samples per frame must be a multiple of encoder.FrameSize
inDataSize = inDataSize - inDataSize % encoder.FrameSize;
var encodedData = new byte[len];
int encodedBytes = encoder.Encode(data, 0, inDataSize, encodedData, 0, len);
if (encodedBytes != 0)
{
// each chunk is laid out as follows:
// | 4-byte total chunk size | 4-byte encoded buffer size | <encoded-bytes> |
byte[] inDataSizeBuf = BitConverter.GetBytes(inDataSize);
byte[] sizeBuf = BitConverter.GetBytes(encodedBytes + inDataSizeBuf.Length);
byte[] returnBuf = new byte[encodedBytes + sizeBuf.Length + inDataSizeBuf.Length];
sizeBuf.CopyTo(returnBuf, 0);
inDataSizeBuf.CopyTo(returnBuf, sizeBuf.Length);
Array.Copy(encodedData, 0, returnBuf, sizeBuf.Length + inDataSizeBuf.Length, encodedBytes);
return returnBuf;
}
else
return buf;
}
void Awake()
{
this.encoder = new SpeexEncoder(bandMode);
this.decoder = new SpeexDecoder(bandMode);
}
public ObservableSpeexEncoder(int bufferSize, BandMode mode)
{
_encoder = new SpeexEncoder(mode);
_encoderInputBuffer = new WaveBuffer(bufferSize);
}
public EncodeSubject(BandMode mode)
{
_encoder = new SpeexEncoder(mode);
}
public void CreateAudioContexts()
{
_captureSource.VideoCaptureDevice = null;
if (_captureSource.AudioCaptureDevice == null)
{
_captureSource.AudioCaptureDevice = CaptureDeviceConfiguration.GetDefaultAudioCaptureDevice();
if (_captureSource.AudioCaptureDevice == null)
{
throw new InvalidOperationException("No suitable audio capture device was found");
}
}
MediaDeviceConfig.SelectBestAudioFormat(_captureSource.AudioCaptureDevice);
_captureSource.AudioCaptureDevice.AudioFrameSize = AudioFormat.Default.MillisecondsPerFrame; // 20 milliseconds
var desiredFormat = _captureSource.AudioCaptureDevice.DesiredFormat;
var rawAudioFormat = new AudioFormat(desiredFormat.SamplesPerSecond, AudioFormat.Default.MillisecondsPerFrame, desiredFormat.Channels, desiredFormat.BitsPerSample);
var playedAudioFormat = new AudioFormat();
var config = MediaConfig.Default;
// Absolutely bare minimum processing - doesn't process sound at all.
var nullAudioFormat = new AudioFormat();
var nullResampler = new ResampleFilter(rawAudioFormat, nullAudioFormat);
nullResampler.InstanceName = "Null resample filter";
var nullEnhancer = new NullEchoCancelFilter(config.ExpectedAudioLatency, config.FilterLength, nullAudioFormat, playedAudioFormat);
nullEnhancer.InstanceName = "Null";
var nullDtx = new NullAudioInplaceFilter();
var nullEncoder = new NullAudioEncoder();
var nullAudioContext = new AudioContext(nullAudioFormat, nullResampler, nullDtx, nullEnhancer, nullEncoder);
nullAudioContext.Description = "Null";
// What we should use when there's only one other person, and CPU is OK:
// 16Khz, Speex, WebRtc at full strength
var directAudioFormat = new AudioFormat();
var directResampler = new ResampleFilter(rawAudioFormat, directAudioFormat);
directResampler.InstanceName = "Direct high quality resample filter";
var directEnhancer = new WebRtcFilter(config.ExpectedAudioLatency, config.FilterLength, directAudioFormat, playedAudioFormat, config.EnableAec, config.EnableDenoise, config.EnableAgc);
directEnhancer.InstanceName = "High";
var directDtx = new DtxFilter(directAudioFormat);
var directEncoder = new SpeexEncoder(directAudioFormat);
var directAudioContext = new AudioContext(directAudioFormat, directResampler, directDtx, directEnhancer, directEncoder);
directAudioContext.Description = "High Quality Direct";
// What we should use when there are multiple people (and hence the audio will need to be decoded and mixed), but CPU is OK:
// 8Khz, G711, WebRtc at full strength
var conferenceAudioFormat = new AudioFormat(AudioConstants.NarrowbandSamplesPerSecond);
var conferenceResampler = new ResampleFilter(rawAudioFormat, conferenceAudioFormat);
conferenceResampler.InstanceName = "Conference high quality resample filter";
var conferenceEnhancer = new WebRtcFilter(config.ExpectedAudioLatency, config.FilterLength, conferenceAudioFormat, playedAudioFormat, config.EnableAec, config.EnableDenoise, config.EnableAgc);
conferenceEnhancer.InstanceName = "Medium";
var conferenceDtx = new DtxFilter(conferenceAudioFormat);
var conferenceEncoder = new G711MuLawEncoder(conferenceAudioFormat);
var conferenceAudioContext = new AudioContext(conferenceAudioFormat, conferenceResampler, conferenceDtx, conferenceEnhancer, conferenceEncoder);
conferenceAudioContext.Description = "High Quality Conference";
// What we should use when one or more remote CPU's isn't keeping up (regardless of how many people are in the room):
// 8Khz, G711, WebRtc at full-strength
var remoteFallbackAudioFormat = new AudioFormat(AudioConstants.NarrowbandSamplesPerSecond);
var remoteFallbackResampler = new ResampleFilter(rawAudioFormat, remoteFallbackAudioFormat);
remoteFallbackResampler.InstanceName = "Fallback remote high cpu resample filter";
var remoteFallbackEnhancer = new WebRtcFilter(config.ExpectedAudioLatency, config.FilterLength, remoteFallbackAudioFormat, playedAudioFormat, config.EnableAec, config.EnableDenoise, config.EnableAgc);
remoteFallbackEnhancer.InstanceName = "Medium";
var remoteFallbackDtx = new DtxFilter(remoteFallbackAudioFormat);
var remoteFallbackEncoder = new G711MuLawEncoder(remoteFallbackAudioFormat);
var remoteFallbackAudioContext = new AudioContext(remoteFallbackAudioFormat, remoteFallbackResampler, remoteFallbackDtx, remoteFallbackEnhancer, remoteFallbackEncoder);
remoteFallbackAudioContext.Description = "Fallback for remote high CPU";
// What we should use when the local CPU isn't keeping up (regardless of how many people are in the room):
// 8Khz, G711, WebRtc at half-strength
var fallbackAudioFormat = new AudioFormat(AudioConstants.NarrowbandSamplesPerSecond);
var fallbackResampler = new ResampleFilter(rawAudioFormat, fallbackAudioFormat);
fallbackResampler.InstanceName = "Fallback resample filter";
var fallbackEnhancer = new WebRtcFilter(config.ExpectedAudioLatencyFallback, config.FilterLengthFallback, fallbackAudioFormat, playedAudioFormat, config.EnableAec, false, false);
fallbackEnhancer.InstanceName = "Low";
var fallbackDtx = new DtxFilter(fallbackAudioFormat);
var fallbackEncoder = new G711MuLawEncoder(fallbackAudioFormat);
var fallbackAudioContext = new AudioContext(fallbackAudioFormat, fallbackResampler, fallbackDtx, fallbackEnhancer, fallbackEncoder);
fallbackAudioContext.Description = "Fallback for local high CPU";
AudioContextCollection.Clear();
AudioContextCollection.Add(nullAudioContext);
AudioContextCollection.Add(directAudioContext);
AudioContextCollection.Add(conferenceAudioContext);
AudioContextCollection.Add(remoteFallbackAudioContext);
AudioContextCollection.Add(fallbackAudioContext);
CurrentAudioContext = nullAudioContext;
}
public void FileInfoTest()
{
const string name = "male.wav";
const int HEAD_SIZE = 8;
const string CHUNK_DATA = "data";
FileInfo file = GetWavFile(name);
Console.WriteLine("file name:" + file.FullName);
FileStream fs = File.OpenRead(file.FullName);
byte[] tmp = new byte[2560];
fs.Read(tmp, 0, HEAD_SIZE + 4);
String ckID = LittleEndian.ReadString(tmp, 0, 4);
Console.WriteLine("ckID,0-3:" + ckID);
int cksize = LittleEndian.ReadInt(tmp, 4);
Console.WriteLine("cksize,4-7:" + cksize);
String WAVEID = LittleEndian.ReadString(tmp, 8, 4);
Console.WriteLine("WAVEID,8-11:" + WAVEID);
fs.Read(tmp, 0, HEAD_SIZE);
String chunk = LittleEndian.ReadString(tmp, 0, 4);
int size = LittleEndian.ReadInt(tmp, 4);
//
short format = 0;
short nChannels = 0;
int nSamplesPerSec = 0;
int nAvgBytesPerSec = 0;
short nBlockAlign = 0;
short wBitsPerSample = 0;
while (!chunk.Equals(CHUNK_DATA))
{
Console.WriteLine("chunk:" + chunk);
Console.WriteLine("size:" + size);
// read size bytes
fs.Read(tmp, 0, size);
format = LittleEndian.ReadShort(tmp, 0);
Console.WriteLine("format,0-1:" + format);
nChannels = LittleEndian.ReadShort(tmp, 2);
Console.WriteLine("nChannels,2-3:" + nChannels);
nSamplesPerSec = LittleEndian.ReadInt(tmp, 4);
Console.WriteLine("nSamplesPerSec,4-7:" + nSamplesPerSec);
nAvgBytesPerSec = LittleEndian.ReadInt(tmp, 8);
Console.WriteLine("nAvgBytesPerSec:" + nAvgBytesPerSec);
nBlockAlign = LittleEndian.ReadShort(tmp, 12);
Console.WriteLine("nBlockAlign:" + nBlockAlign);
wBitsPerSample = LittleEndian.ReadShort(tmp, 14);
Console.WriteLine("wBitsPerSample:" + wBitsPerSample);
fs.Read(tmp, 0, HEAD_SIZE);
chunk = LittleEndian.ReadString(tmp, 0, 4);
size = LittleEndian.ReadInt(tmp, 4);
Console.WriteLine("chunk:" + chunk);
Console.WriteLine("size:" + size);
}
BandMode mode;
if (nSamplesPerSec < 12000)
{
// Narrowband
mode = BandMode.Narrow;
}
else if (nSamplesPerSec < 24000)
{
// Wideband
mode = BandMode.Wide;
}
else
{
//Ultra-wideband
mode = BandMode.UltraWide;
}
SpeexEncoder speexEncoder = new SpeexEncoder(mode);
int quality = 2; // modify your self
int framesPerPackage = 4; // one frame one package
bool vbr = false;
// speexEncoder.getEncoder().setComplexity(5);
// speexEncoder.getEncoder().setBitRate();
// speexEncoder.getEncoder().setVbr();
// speexEncoder.getEncoder().setVbrQuality();
// speexEncoder.getEncoder().setVad();
// speexEncoder.getEncoder().setDtx();
OggSpeexWriter2 writer = new OggSpeexWriter2(mode, nSamplesPerSec, nChannels, framesPerPackage, vbr, nAvgBytesPerSec);
writer.Open("result.spx");
writer.WriteHeader("alking");
int pcmPacketSize = 2 * nChannels * speexEncoder.FrameSize;
int bytesCount = 0;
while (bytesCount < size)
{
int read = pcmPacketSize * framesPerPackage;
fs.Read(tmp, 0, read);
short[] data = new short[read / 2];
Buffer.BlockCopy(tmp, 0, data, 0, read);
int encSize = speexEncoder.Encode(data, 0, data.Length, tmp, 0, tmp.Length);
if (encSize > 0)
{
writer.WritePackage(tmp, 0, encSize);
}
bytesCount += read;
}
writer.Close();
fs.Close();
}
void Awake()
{
this.encoder = new SpeexEncoder(bandMode);
this.decoder = new SpeexDecoder(bandMode);
}
/// <summary>
/// 初始化。
/// </summary>
public Speex()
{
_encoder = new SpeexEncoder(BandMode.Narrow);
_decoder = new SpeexDecoder(BandMode.Narrow);
}
