ioquake
/
ioq3
mirror of https://github.com/ioquake/ioq3.git
0
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

VoIP: attempt at changing voipPower calc and using it for VAD. 

The Speex VAD sort of sucks, honestly, or I'm not using it right. Now
 trying this algorithm, after denoising:

    http://lists.xiph.org/pipermail/speex-dev/2006-March/004269.html

And I'll play around to find the threshold for considering a set of frames
 to be "voice" from there.

Also worth noting: we consider the power of the set of frames as a whole, so
 you need to sustain power for 0.25 seconds at a time, or it's not "voice."
This commit is contained in:
Ryan C. Gordon 2008-06-08 07:09:10 +00:00
parent fdfaff64f8
commit d4ee08b985
2 changed files with 12 additions and 22 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
code/client/cl_cgame.c
View File

@ -937,10 +937,6 @@ void CL_FirstSnapshot( void ) {
speex_preprocess_ctl(clc.speexPreprocessor,
SPEEX_PREPROCESS_SET_DENOISE, &i);
i = (cl_voipUseVAD->integer != 0);
speex_preprocess_ctl(clc.speexPreprocessor,
SPEEX_PREPROCESS_SET_VAD, &i);
for (i = 0; i < MAX_CLIENTS; i++) {
speex_bits_init(&clc.speexDecoderBits[i]);
speex_bits_reset(&clc.speexDecoderBits[i]);

30
code/client/cl_main.c
View File

@ -283,11 +283,8 @@ void CL_CaptureVoip(void)
return; // packet is pending transmission, don't record more yet.
if (cl_voipUseVAD->modified) {
int useVadi = (int) useVad;
speex_preprocess_ctl(clc.speexPreprocessor,
SPEEX_PREPROCESS_SET_VAD, &useVadi);
cl_voipUseVAD->modified = qfalse;
Cvar_Set("cl_voipSend", (useVad) ? "1" : "0");
cl_voipUseVAD->modified = qfalse;
}
if ((useVad) && (!cl_voipSend->integer))
@ -339,8 +336,7 @@ void CL_CaptureVoip(void)
// audio capture is always MONO16 (and that's what speex wants!).
// 2048 will cover 12 uncompressed frames in narrowband mode.
static int16_t sampbuffer[2048];
qboolean isVoice = qfalse;
int16_t voipPower = 0;
float voipPower = 0.0f;
int speexFrames = 0;
int wpos = 0;
int pos = 0;
@ -359,19 +355,15 @@ void CL_CaptureVoip(void)
int16_t *sampptr = &sampbuffer[pos];
int i, bytes;
// preprocess samples to remove noise...
speex_preprocess_run(clc.speexPreprocessor, sampptr);
// check the "power" of this packet...
for (i = 0; i < clc.speexFrameSize; i++) {
int16_t s = sampptr[i];
if (s < 0)
s = -s;
if (s > voipPower)
voipPower = s; // !!! FIXME: this isn't very clever.
const float s = fabs((float) sampptr[i]);
voipPower += s * s;
}
// preprocess samples to remove noise, check for voice...
if (speex_preprocess_run(clc.speexPreprocessor, sampptr))
isVoice = qtrue; // player is probably speaking.
// encode raw audio samples into Speex data...
speex_bits_reset(&clc.speexEncoderBits);
speex_encode_int(clc.speexEncoder, sampptr,
@ -389,10 +381,12 @@ void CL_CaptureVoip(void)
speexFrames++;
}
if ((useVad) && (!isVoice)) {
CL_VoipNewGeneration(); // no talk for at least 1/4 second.
clc.voipPower = voipPower / (32768.0f * 32768.0f *
((float) (clc.speexFrameSize * speexFrames)));
if ((useVad) && (clc.voipPower > 0.25f)) {
CL_VoipNewGeneration(); // no "talk" for at least 1/4 second.
} else {
clc.voipPower = ((float) voipPower) / 32767.0f;
clc.voipOutgoingDataSize = wpos;
clc.voipOutgoingDataFrames = speexFrames;