source-engine/engine/audio/private/voice.cpp
2022-07-28 18:32:42 +03:00

1579 lines
42 KiB
C++

//========= Copyright Valve Corporation, All rights reserved. ============//
//
// Purpose:
//
// $NoKeywords: $
//
//=============================================================================//
#include "audio_pch.h"
#include "circularbuffer.h"
#include "voice.h"
#include "voice_wavefile.h"
#include "r_efx.h"
#include "cdll_int.h"
#include "voice_gain.h"
#include "voice_mixer_controls.h"
#include "ivoicerecord.h"
#include "ivoicecodec.h"
#include "filesystem.h"
#include "../../filesystem_engine.h"
#include "tier1/utlbuffer.h"
#if defined( _X360 )
#include "xauddefs.h"
#endif
#include "steam/steam_api.h"
// memdbgon must be the last include file in a .cpp file!!!
#include "tier0/memdbgon.h"
static CSteamAPIContext g_SteamAPIContext;
static CSteamAPIContext *steamapicontext = NULL;
void Voice_EndChannel( int iChannel );
void VoiceTweak_EndVoiceTweakMode();
void EngineTool_OverrideSampleRate( int& rate );
// A fallback codec that should be the most likely to work for local/offline use
#define VOICE_FALLBACK_CODEC "vaudio_opus"
// Special entity index used for tweak mode.
#define TWEAKMODE_ENTITYINDEX -500
// Special channel index passed to Voice_AddIncomingData when in tweak mode.
#define TWEAKMODE_CHANNELINDEX -100
// How long does the sign stay above someone's head when they talk?
#define SPARK_TIME 0.2
// How long a voice channel has to be inactive before we free it.
#define DIE_COUNTDOWN 0.5
#define VOICE_RECEIVE_BUFFER_SIZE (VOICE_OUTPUT_SAMPLE_RATE_MAX * BYTES_PER_SAMPLE)
#define LOCALPLAYERTALKING_TIMEOUT 0.2f // How long it takes for the client to decide the server isn't sending acks
// of voice data back.
// If this is defined, then the data is converted to 8-bit and sent otherwise uncompressed.
// #define VOICE_SEND_RAW_TEST
// The format we sample voice in.
WAVEFORMATEX g_VoiceSampleFormat =
{
WAVE_FORMAT_PCM, // wFormatTag
1, // nChannels
// These two can be dynamically changed by voice_init
VOICE_OUTPUT_SAMPLE_RATE_LOW, // nSamplesPerSec
VOICE_OUTPUT_SAMPLE_RATE_LOW*2, // nAvgBytesPerSec
2, // nBlockAlign
16, // wBitsPerSample
sizeof(WAVEFORMATEX) // cbSize
};
static bool Voice_SetSampleRate( DWORD rate )
{
if ( g_VoiceSampleFormat.nSamplesPerSec != rate ||
g_VoiceSampleFormat.nAvgBytesPerSec != rate * 2 )
{
g_VoiceSampleFormat.nSamplesPerSec = rate;
g_VoiceSampleFormat.nAvgBytesPerSec = rate * 2;
return true;
}
return false;
}
int Voice_SamplesPerSec()
{
int rate = g_VoiceSampleFormat.nSamplesPerSec;
EngineTool_OverrideSampleRate( rate );
return rate;
}
int Voice_AvgBytesPerSec()
{
int rate = g_VoiceSampleFormat.nSamplesPerSec;
EngineTool_OverrideSampleRate( rate );
return ( rate * g_VoiceSampleFormat.wBitsPerSample ) >> 3;
}
ConVar voice_avggain( "voice_avggain", "0.5" );
ConVar voice_maxgain( "voice_maxgain", "10" );
ConVar voice_scale( "voice_scale", "1", FCVAR_ARCHIVE );
ConVar voice_loopback( "voice_loopback", "0", FCVAR_USERINFO );
ConVar voice_fadeouttime( "voice_fadeouttime", "0.1" ); // It fades to no sound at the tail end of your voice data when you release the key.
// Debugging cvars.
ConVar voice_profile( "voice_profile", "0" );
ConVar voice_showchannels( "voice_showchannels", "0" ); // 1 = list channels
// 2 = show timing info, etc
ConVar voice_showincoming( "voice_showincoming", "0" ); // show incoming voice data
ConVar voice_enable( "voice_enable", "1", FCVAR_ARCHIVE ); // Globally enable or disable voice.
#ifdef VOICE_VOX_ENABLE
ConVar voice_threshold( "voice_threshold", "2000", FCVAR_ARCHIVE );
#endif // VOICE_VOX_ENABLE
// Have it force your mixer control settings so waveIn comes from the microphone.
// CD rippers change your waveIn to come from the CD drive
ConVar voice_forcemicrecord( "voice_forcemicrecord", "1", FCVAR_ARCHIVE );
// This should not be lower than the maximum difference between clients' frame durations (due to cmdrate/updaterate),
// plus some jitter allowance.
ConVar voice_buffer_ms( "voice_buffer_ms", "100", FCVAR_INTERNAL_USE,
"How many milliseconds of voice to buffer to avoid dropouts due to jitter and frame time differences." );
int g_nVoiceFadeSamples = 1; // Calculated each frame from the cvar.
float g_VoiceFadeMul = 1; // 1 / (g_nVoiceFadeSamples - 1).
// While in tweak mode, you can't hear anything anyone else is saying, and your own voice data
// goes directly to the speakers.
bool g_bInTweakMode = false;
int g_VoiceTweakSpeakingVolume = 0;
bool g_bVoiceAtLeastPartiallyInitted = false;
// The codec and sample rate passed to Voice_Init. "" and -1 if voice is not initialized
char g_szVoiceCodec[_MAX_PATH] = { 0 };
int g_nVoiceRequestedSampleRate = -1;
const char *Voice_ConfiguredCodec() { return g_szVoiceCodec; }
int Voice_ConfiguredSampleRate() { return g_nVoiceRequestedSampleRate; }
// Timing info for each frame.
static double g_CompressTime = 0;
static double g_DecompressTime = 0;
static double g_GainTime = 0;
static double g_UpsampleTime = 0;
class CVoiceTimer
{
public:
inline void Start()
{
if( voice_profile.GetInt() )
{
m_StartTime = Plat_FloatTime();
}
}
inline void End(double *out)
{
if( voice_profile.GetInt() )
{
*out += Plat_FloatTime() - m_StartTime;
}
}
double m_StartTime;
};
static bool g_bLocalPlayerTalkingAck = false;
static float g_LocalPlayerTalkingTimeout = 0;
CSysModule *g_hVoiceCodecDLL = 0;
// Voice recorder. Can be waveIn, DSound, or whatever.
static IVoiceRecord *g_pVoiceRecord = NULL;
static IVoiceCodec *g_pEncodeCodec = NULL;
static bool g_bVoiceRecording = false; // Are we recording at the moment?
static bool g_bVoiceRecordStopping = false; // Are we waiting to stop?
bool g_bUsingSteamVoice = false;
#ifdef WIN32
extern IVoiceRecord* CreateVoiceRecord_DSound(int nSamplesPerSec);
#elif defined( OSX )
extern IVoiceRecord* CreateVoiceRecord_AudioQueue(int sampleRate);
#endif
#ifdef POSIX
extern IVoiceRecord* CreateVoiceRecord_OpenAL(int sampleRate);
#endif
#ifdef USE_SDL
extern IVoiceRecord *CreateVoiceRecord_SDL(int sampleRate);
#endif
static bool VoiceRecord_Start()
{
if ( g_bUsingSteamVoice )
{
if ( steamapicontext && steamapicontext->SteamUser() )
{
steamapicontext->SteamUser()->StartVoiceRecording();
return true;
}
}
else if ( g_pVoiceRecord )
{
return g_pVoiceRecord->RecordStart();
}
return false;
}
static void VoiceRecord_Stop()
{
if ( g_bUsingSteamVoice )
{
if ( steamapicontext && steamapicontext->SteamUser() )
{
steamapicontext->SteamUser()->StopVoiceRecording();
}
}
else if ( g_pVoiceRecord )
{
return g_pVoiceRecord->RecordStop();
}
}
//
// Used for storing incoming voice data from an entity.
//
class CVoiceChannel
{
public:
CVoiceChannel();
// Called when someone speaks and a new voice channel is setup to hold the data.
void Init(int nEntity);
public:
int m_iEntity; // Number of the entity speaking on this channel (index into cl_entities).
// This is -1 when the channel is unused.
CSizedCircularBuffer
<VOICE_RECEIVE_BUFFER_SIZE> m_Buffer; // Circular buffer containing the voice data.
// Used for upsampling..
double m_LastFraction; // Fraction between m_LastSample and the next sample.
short m_LastSample;
bool m_bStarved; // Set to true when the channel runs out of data for the mixer.
// The channel is killed at that point.
float m_TimePad; // Set to TIME_PADDING when the first voice packet comes in from a sender.
// We add time padding (for frametime differences)
// by waiting a certain amount of time before starting to output the sound.
IVoiceCodec *m_pVoiceCodec; // Each channel gets is own IVoiceCodec instance so the codec can maintain state.
CAutoGain m_AutoGain; // Gain we're applying to this channel
CVoiceChannel *m_pNext;
bool m_bProximity;
int m_nViewEntityIndex;
int m_nSoundGuid;
};
CVoiceChannel::CVoiceChannel()
{
m_iEntity = -1;
m_pVoiceCodec = NULL;
m_nViewEntityIndex = -1;
m_nSoundGuid = -1;
}
void CVoiceChannel::Init(int nEntity)
{
m_iEntity = nEntity;
m_bStarved = false;
m_Buffer.Flush();
m_TimePad = Clamp( voice_buffer_ms.GetFloat(), 1.f, 5000.f ) / 1000.f;
m_LastSample = 0;
m_LastFraction = 0.999;
m_AutoGain.Reset( 128, voice_maxgain.GetFloat(), voice_avggain.GetFloat(), voice_scale.GetFloat() );
}
// Incoming voice channels.
CVoiceChannel g_VoiceChannels[VOICE_NUM_CHANNELS];
// These are used for recording the wave data into files for debugging.
#define MAX_WAVEFILEDATA_LEN 1024*1024
char *g_pUncompressedFileData = NULL;
int g_nUncompressedDataBytes = 0;
const char *g_pUncompressedDataFilename = NULL;
char *g_pDecompressedFileData = NULL;
int g_nDecompressedDataBytes = 0;
const char *g_pDecompressedDataFilename = NULL;
char *g_pMicInputFileData = NULL;
int g_nMicInputFileBytes = 0;
int g_CurMicInputFileByte = 0;
double g_MicStartTime;
static ConVar voice_writevoices( "voice_writevoices", "0", 0, "Saves each speaker's voice data into separate .wav files\n" );
class CVoiceWriterData
{
public:
CVoiceWriterData() :
m_pChannel( NULL ),
m_nCount( 0 ),
m_Buffer()
{
}
// Copy ctor is needed to insert into CVoiceWriter's CUtlRBTree.
CVoiceWriterData(const CVoiceWriterData& other) :
m_pChannel( other.m_pChannel ),
m_nCount( other.m_nCount ),
m_Buffer( )
{
m_Buffer.CopyBuffer( other.m_Buffer );
}
static bool Less( const CVoiceWriterData &lhs, const CVoiceWriterData &rhs )
{
return lhs.m_pChannel < rhs.m_pChannel;
}
CVoiceChannel *m_pChannel;
int m_nCount;
CUtlBuffer m_Buffer;
private:
CVoiceWriterData& operator=(const CVoiceWriterData&);
};
class CVoiceWriter
{
public:
CVoiceWriter() :
m_VoiceWriter( 0, 0, CVoiceWriterData::Less )
{
}
void Flush()
{
for ( int i = m_VoiceWriter.FirstInorder(); i != m_VoiceWriter.InvalidIndex(); i = m_VoiceWriter.NextInorder( i ) )
{
CVoiceWriterData *data = &m_VoiceWriter[ i ];
if ( data->m_Buffer.TellPut() <= 0 )
continue;
data->m_Buffer.Purge();
}
}
void Finish()
{
if ( !g_pSoundServices->IsConnected() )
{
Flush();
return;
}
for ( int i = m_VoiceWriter.FirstInorder(); i != m_VoiceWriter.InvalidIndex(); i = m_VoiceWriter.NextInorder( i ) )
{
CVoiceWriterData *data = &m_VoiceWriter[ i ];
if ( data->m_Buffer.TellPut() <= 0 )
continue;
int index = data->m_pChannel - g_VoiceChannels;
Assert( index >= 0 && index < (int)ARRAYSIZE( g_VoiceChannels ) );
char path[ MAX_PATH ];
Q_snprintf( path, sizeof( path ), "%s/voice", g_pSoundServices->GetGameDir() );
g_pFileSystem->CreateDirHierarchy( path );
char fn[ MAX_PATH ];
Q_snprintf( fn, sizeof( fn ), "%s/pl%02d_slot%d-time%d.wav", path, index, data->m_nCount, (int)g_pSoundServices->GetClientTime() );
WriteWaveFile( fn, (const char *)data->m_Buffer.Base(), data->m_Buffer.TellPut(), g_VoiceSampleFormat.wBitsPerSample, g_VoiceSampleFormat.nChannels, Voice_SamplesPerSec() );
Msg( "Writing file %s\n", fn );
++data->m_nCount;
data->m_Buffer.Purge();
}
}
void AddDecompressedData( CVoiceChannel *ch, const byte *data, size_t datalen )
{
if ( !voice_writevoices.GetBool() )
return;
CVoiceWriterData search;
search.m_pChannel = ch;
int idx = m_VoiceWriter.Find( search );
if ( idx == m_VoiceWriter.InvalidIndex() )
{
idx = m_VoiceWriter.Insert( search );
}
CVoiceWriterData *slot = &m_VoiceWriter[ idx ];
slot->m_Buffer.Put( data, datalen );
}
private:
CUtlRBTree< CVoiceWriterData > m_VoiceWriter;
};
static CVoiceWriter g_VoiceWriter;
inline void ApplyFadeToSamples(short *pSamples, int nSamples, int fadeOffset, float fadeMul)
{
for(int i=0; i < nSamples; i++)
{
float percent = (i+fadeOffset) * fadeMul;
pSamples[i] = (short)(pSamples[i] * (1 - percent));
}
}
bool Voice_Enabled( void )
{
return voice_enable.GetBool();
}
int Voice_GetOutputData(
const int iChannel, //! The voice channel it wants samples from.
char *copyBufBytes, //! The buffer to copy the samples into.
const int copyBufSize, //! Maximum size of copyBuf.
const int samplePosition, //! Which sample to start at.
const int sampleCount //! How many samples to get.
)
{
CVoiceChannel *pChannel = &g_VoiceChannels[iChannel];
short *pCopyBuf = (short*)copyBufBytes;
int maxOutSamples = copyBufSize / BYTES_PER_SAMPLE;
// Find out how much we want and get it from the received data channel.
CCircularBuffer *pBuffer = &pChannel->m_Buffer;
int nBytesToRead = pBuffer->GetReadAvailable();
nBytesToRead = min(min(nBytesToRead, (int)maxOutSamples), sampleCount * BYTES_PER_SAMPLE);
int nSamplesGotten = pBuffer->Read(pCopyBuf, nBytesToRead) / BYTES_PER_SAMPLE;
// Are we at the end of the buffer's data? If so, fade data to silence so it doesn't clip.
int readSamplesAvail = pBuffer->GetReadAvailable() / BYTES_PER_SAMPLE;
if(readSamplesAvail < g_nVoiceFadeSamples)
{
int bufferFadeOffset = max((readSamplesAvail + nSamplesGotten) - g_nVoiceFadeSamples, 0);
int globalFadeOffset = max(g_nVoiceFadeSamples - (readSamplesAvail + nSamplesGotten), 0);
ApplyFadeToSamples(
&pCopyBuf[bufferFadeOffset],
nSamplesGotten - bufferFadeOffset,
globalFadeOffset,
g_VoiceFadeMul);
}
// If there weren't enough samples in the received data channel,
// pad it with a copy of the most recent data, and if there
// isn't any, then use zeros.
if ( nSamplesGotten < sampleCount )
{
int wantedSampleCount = min( sampleCount, maxOutSamples );
int nAdditionalNeeded = (wantedSampleCount - nSamplesGotten);
if ( nSamplesGotten > 0 )
{
short *dest = (short *)&pCopyBuf[ nSamplesGotten ];
int nSamplesToDuplicate = min( nSamplesGotten, nAdditionalNeeded );
const short *src = (short *)&pCopyBuf[ nSamplesGotten - nSamplesToDuplicate ];
Q_memcpy( dest, src, nSamplesToDuplicate * BYTES_PER_SAMPLE );
//Msg( "duplicating %d samples\n", nSamplesToDuplicate );
nAdditionalNeeded -= nSamplesToDuplicate;
if ( nAdditionalNeeded > 0 )
{
dest = (short *)&pCopyBuf[ nSamplesGotten + nSamplesToDuplicate ];
Q_memset(dest, 0, nAdditionalNeeded * BYTES_PER_SAMPLE);
// Msg( "zeroing %d samples\n", nAdditionalNeeded );
Assert( ( nAdditionalNeeded + nSamplesGotten + nSamplesToDuplicate ) == wantedSampleCount );
}
}
else
{
Q_memset( &pCopyBuf[ nSamplesGotten ], 0, nAdditionalNeeded * BYTES_PER_SAMPLE );
}
nSamplesGotten = wantedSampleCount;
}
// If the buffer is out of data, mark this channel to go away.
if(pBuffer->GetReadAvailable() == 0)
{
pChannel->m_bStarved = true;
}
if(voice_showchannels.GetInt() >= 2)
{
Msg("Voice - mixed %d samples from channel %d\n", nSamplesGotten, iChannel);
}
VoiceSE_MoveMouth(pChannel->m_iEntity, (short*)copyBufBytes, nSamplesGotten);
return nSamplesGotten;
}
void Voice_OnAudioSourceShutdown( int iChannel )
{
Voice_EndChannel( iChannel );
}
// ------------------------------------------------------------------------ //
// Internal stuff.
// ------------------------------------------------------------------------ //
CVoiceChannel* GetVoiceChannel(int iChannel, bool bAssert=true)
{
if(iChannel < 0 || iChannel >= VOICE_NUM_CHANNELS)
{
if(bAssert)
{
Assert(false);
}
return NULL;
}
else
{
return &g_VoiceChannels[iChannel];
}
}
// Helper for doing a default-init with some codec if we weren't passed specific parameters
bool Voice_InitWithDefault( const char *pCodecName )
{
if ( !pCodecName || !*pCodecName )
{
return false;
}
int nRate = Voice_GetDefaultSampleRate( pCodecName );
if ( nRate < 0 )
{
Msg( "Voice_InitWithDefault: Unable to determine defaults for codec \"%s\"\n", pCodecName );
return false;
}
return Voice_Init( pCodecName, Voice_GetDefaultSampleRate( pCodecName ) );
}
bool Voice_Init( const char *pCodecName, int nSampleRate )
{
if ( voice_enable.GetInt() == 0 )
{
return false;
}
if ( !pCodecName || !pCodecName[0] )
{
return false;
}
bool bSpeex = Q_stricmp( pCodecName, "vaudio_speex" ) == 0;
bool bCelt = Q_stricmp( pCodecName, "vaudio_celt" ) == 0;
bool bOpus = Q_stricmp( pCodecName, "vaudio_opus" ) == 0;
bool bSteam = Q_stricmp( pCodecName, "steam" ) == 0;
// Miles has not been in use for voice in a long long time. Not worth the surface to support ancient demos that may
// use it (and probably do not work for other reasons)
// "vaudio_miles"
if ( !( bSpeex || bCelt || bOpus || bSteam ) )
{
Msg( "Voice_Init Failed: invalid voice codec %s.\n", pCodecName );
return false;
}
Voice_Deinit();
g_bVoiceAtLeastPartiallyInitted = true;
V_strncpy( g_szVoiceCodec, pCodecName, sizeof(g_szVoiceCodec) );
g_nVoiceRequestedSampleRate = nSampleRate;
g_bUsingSteamVoice = bSteam;
if ( !steamapicontext )
{
steamapicontext = &g_SteamAPIContext;
steamapicontext->Init();
}
if ( g_bUsingSteamVoice )
{
if ( !steamapicontext->SteamFriends() || !steamapicontext->SteamUser() )
{
Msg( "Voice_Init: Requested Steam voice, but cannot access API. Voice will not function\n" );
return false;
}
}
// For steam, nSampleRate 0 means "use optimal steam sample rate".
if ( bSteam && nSampleRate == 0 )
{
Msg( "Voice_Init: Using Steam voice optimal sample rate %d\n",
steamapicontext->SteamUser()->GetVoiceOptimalSampleRate() );
// Steam's sample rate may change and not be supported by our rather unflexible sound engine. However, steam
// will resample as necessary in DecompressVoice, so we can pretend we're outputting at native rates.
//
// Behind the scenes, we'll request steam give us the encoded stream at its "optimal" rate, then we'll try to
// decompress the output at this rate, making it transparent to us that the encoded stream is not at our output
// rate.
Voice_SetSampleRate( SOUND_DMA_SPEED );
}
else
{
Voice_SetSampleRate( nSampleRate );
}
if(!VoiceSE_Init())
return false;
// Get the voice input device.
#ifdef OSX
g_pVoiceRecord = CreateVoiceRecord_AudioQueue( Voice_SamplesPerSec() );
if ( !g_pVoiceRecord )
{
// Fall back to OpenAL
g_pVoiceRecord = CreateVoiceRecord_OpenAL( Voice_SamplesPerSec() );
}
#elif defined( WIN32 )
g_pVoiceRecord = CreateVoiceRecord_DSound( Voice_SamplesPerSec() );
#elif defined( USE_SDL )
g_pVoiceRecord = CreateVoiceRecord_SDL( Voice_SamplesPerSec() );
#else
g_pVoiceRecord = CreateVoiceRecord_OpenAL( Voice_SamplesPerSec() );
#endif
if( !g_pVoiceRecord )
{
Msg( "Unable to initialize sound capture. You won't be able to speak to other players." );
}
// Init codec DLL for non-steam
if ( !bSteam )
{
// CELT's qualities are 0-3, we historically just passed 4 to the other two even though they don't really map to the
// same thing.
//
// Changing the quality level we use here will require either extending SVC_VoiceInit to pass down which quality is
// in use or using a different codec name (vaudio_celtHD!) for backwards compatibility
int quality = ( bCelt || bOpus ) ? 3 : 4;
// Get the codec.
CreateInterfaceFn createCodecFn = NULL;
g_hVoiceCodecDLL = FileSystem_LoadModule(pCodecName);
if( !g_hVoiceCodecDLL || (createCodecFn = Sys_GetFactory(g_hVoiceCodecDLL)) == NULL )
{
g_hVoiceCodecDLL = FileSystem_LoadModule( VOICE_FALLBACK_CODEC );
pCodecName = VOICE_FALLBACK_CODEC;
}
if ( !g_hVoiceCodecDLL || (createCodecFn = Sys_GetFactory(g_hVoiceCodecDLL)) == NULL ||
(g_pEncodeCodec = (IVoiceCodec*)createCodecFn(pCodecName, NULL)) == NULL || !g_pEncodeCodec->Init( quality ) )
{
Msg("Unable to load voice codec '%s'. Voice disabled. (module %i, iface %i, codec %i)\n",
pCodecName, !!g_hVoiceCodecDLL, !!createCodecFn, !!g_pEncodeCodec);
Voice_Deinit();
return false;
}
for (int i=0; i < VOICE_NUM_CHANNELS; i++)
{
CVoiceChannel *pChannel = &g_VoiceChannels[i];
if ((pChannel->m_pVoiceCodec = (IVoiceCodec*)createCodecFn(pCodecName, NULL)) == NULL || !pChannel->m_pVoiceCodec->Init( quality ))
{
Voice_Deinit();
return false;
}
}
}
// XXX(JohnS): These don't do much in Steam codec mode, but code below uses their presence to mean 'voice fully
// initialized' and other things assume they will succeed.
InitMixerControls();
// Steam mode uses steam for raw input so this isn't meaningful and could have side-effects
if( voice_forcemicrecord.GetInt() && !bSteam )
{
if( g_pMixerControls )
g_pMixerControls->SelectMicrophoneForWaveInput();
}
return true;
}
void Voice_EndChannel(int iChannel)
{
Assert(iChannel >= 0 && iChannel < VOICE_NUM_CHANNELS);
CVoiceChannel *pChannel = &g_VoiceChannels[iChannel];
if ( pChannel->m_iEntity != -1 )
{
int iEnt = pChannel->m_iEntity;
pChannel->m_iEntity = -1;
if ( pChannel->m_bProximity == true )
{
VoiceSE_EndChannel( iChannel, iEnt );
}
else
{
VoiceSE_EndChannel( iChannel, pChannel->m_nViewEntityIndex );
}
g_pSoundServices->OnChangeVoiceStatus( iEnt, false );
VoiceSE_CloseMouth( iEnt );
pChannel->m_nViewEntityIndex = -1;
pChannel->m_nSoundGuid = -1;
// If the tweak mode channel is ending
if ( iChannel == 0 &&
g_bInTweakMode )
{
VoiceTweak_EndVoiceTweakMode();
}
}
}
void Voice_EndAllChannels()
{
for(int i=0; i < VOICE_NUM_CHANNELS; i++)
{
Voice_EndChannel(i);
}
}
bool EngineTool_SuppressDeInit();
void Voice_Deinit()
{
// This call tends to be expensive and when voice is not enabled it will continually
// call in here, so avoid the work if possible.
if( !g_bVoiceAtLeastPartiallyInitted )
return;
if ( EngineTool_SuppressDeInit() )
return;
Voice_EndAllChannels();
Voice_RecordStop();
for(int i=0; i < VOICE_NUM_CHANNELS; i++)
{
CVoiceChannel *pChannel = &g_VoiceChannels[i];
if ( pChannel->m_pVoiceCodec )
{
pChannel->m_pVoiceCodec->Release();
pChannel->m_pVoiceCodec = NULL;
}
}
if( g_pEncodeCodec )
{
g_pEncodeCodec->Release();
g_pEncodeCodec = NULL;
}
if(g_hVoiceCodecDLL)
{
FileSystem_UnloadModule(g_hVoiceCodecDLL);
g_hVoiceCodecDLL = NULL;
}
if(g_pVoiceRecord)
{
g_pVoiceRecord->Release();
g_pVoiceRecord = NULL;
}
VoiceSE_Term();
g_bVoiceAtLeastPartiallyInitted = false;
g_szVoiceCodec[0] = '\0';
g_nVoiceRequestedSampleRate = -1;
g_bUsingSteamVoice = false;
}
bool Voice_GetLoopback()
{
return !!voice_loopback.GetInt();
}
void Voice_LocalPlayerTalkingAck()
{
if(!g_bLocalPlayerTalkingAck)
{
// Tell the client DLL when this changes.
g_pSoundServices->OnChangeVoiceStatus(-2, TRUE);
}
g_bLocalPlayerTalkingAck = true;
g_LocalPlayerTalkingTimeout = 0;
}
void Voice_UpdateVoiceTweakMode()
{
if(!g_bInTweakMode || !g_pVoiceRecord)
return;
CVoiceChannel *pTweakChannel = GetVoiceChannel( 0 );
if ( pTweakChannel->m_nSoundGuid != -1 &&
!S_IsSoundStillPlaying( pTweakChannel->m_nSoundGuid ) )
{
VoiceTweak_EndVoiceTweakMode();
return;
}
char uchVoiceData[4096];
bool bFinal = false;
int nDataLength = Voice_GetCompressedData(uchVoiceData, sizeof(uchVoiceData), bFinal);
Voice_AddIncomingData(TWEAKMODE_CHANNELINDEX, uchVoiceData, nDataLength, 0);
}
void Voice_Idle(float frametime)
{
if( voice_enable.GetInt() == 0 )
{
Voice_Deinit();
return;
}
if( g_bLocalPlayerTalkingAck )
{
g_LocalPlayerTalkingTimeout += frametime;
if(g_LocalPlayerTalkingTimeout > LOCALPLAYERTALKING_TIMEOUT)
{
g_bLocalPlayerTalkingAck = false;
// Tell the client DLL.
g_pSoundServices->OnChangeVoiceStatus(-2, FALSE);
}
}
// Precalculate these to speedup the voice fadeout.
g_nVoiceFadeSamples = max((int)(voice_fadeouttime.GetFloat() * g_VoiceSampleFormat.nSamplesPerSec ), 2);
g_VoiceFadeMul = 1.0f / (g_nVoiceFadeSamples - 1);
if(g_pVoiceRecord)
g_pVoiceRecord->Idle();
// If we're in voice tweak mode, feed our own data back to us.
Voice_UpdateVoiceTweakMode();
// Age the channels.
int nActive = 0;
for(int i=0; i < VOICE_NUM_CHANNELS; i++)
{
CVoiceChannel *pChannel = &g_VoiceChannels[i];
if(pChannel->m_iEntity != -1)
{
if(pChannel->m_bStarved)
{
// Kill the channel. It's done playing.
Voice_EndChannel(i);
pChannel->m_nSoundGuid = -1;
}
else
{
float oldpad = pChannel->m_TimePad;
pChannel->m_TimePad -= frametime;
if(oldpad > 0 && pChannel->m_TimePad <= 0)
{
// Start its audio.
pChannel->m_nViewEntityIndex = g_pSoundServices->GetViewEntity();
pChannel->m_nSoundGuid = VoiceSE_StartChannel( i, pChannel->m_iEntity, pChannel->m_bProximity, pChannel->m_nViewEntityIndex );
g_pSoundServices->OnChangeVoiceStatus(pChannel->m_iEntity, TRUE);
VoiceSE_InitMouth(pChannel->m_iEntity);
}
++nActive;
}
}
}
if(nActive == 0)
VoiceSE_EndOverdrive();
VoiceSE_Idle(frametime);
// voice_showchannels.
if( voice_showchannels.GetInt() >= 1 )
{
for(int i=0; i < VOICE_NUM_CHANNELS; i++)
{
CVoiceChannel *pChannel = &g_VoiceChannels[i];
if(pChannel->m_iEntity == -1)
continue;
Msg("Voice - chan %d, ent %d, bufsize: %d\n", i, pChannel->m_iEntity, pChannel->m_Buffer.GetReadAvailable());
}
}
// Show profiling data?
if( voice_profile.GetInt() )
{
Msg("Voice - compress: %7.2fu, decompress: %7.2fu, gain: %7.2fu, upsample: %7.2fu, total: %7.2fu\n",
g_CompressTime*1000000.0,
g_DecompressTime*1000000.0,
g_GainTime*1000000.0,
g_UpsampleTime*1000000.0,
(g_CompressTime+g_DecompressTime+g_GainTime+g_UpsampleTime)*1000000.0
);
g_CompressTime = g_DecompressTime = g_GainTime = g_UpsampleTime = 0;
}
}
bool Voice_IsRecording()
{
return g_bVoiceRecording && !g_bInTweakMode;
}
bool Voice_RecordStart(
const char *pUncompressedFile,
const char *pDecompressedFile,
const char *pMicInputFile)
{
if( !g_pEncodeCodec && !g_bUsingSteamVoice )
return false;
g_VoiceWriter.Flush();
Voice_RecordStop();
if ( !g_bUsingSteamVoice )
{
g_pEncodeCodec->ResetState();
}
if(pMicInputFile)
{
int a, b, c;
ReadWaveFile(pMicInputFile, g_pMicInputFileData, g_nMicInputFileBytes, a, b, c);
g_CurMicInputFileByte = 0;
g_MicStartTime = Plat_FloatTime();
}
if(pUncompressedFile)
{
g_pUncompressedFileData = new char[MAX_WAVEFILEDATA_LEN];
g_nUncompressedDataBytes = 0;
g_pUncompressedDataFilename = pUncompressedFile;
}
if(pDecompressedFile)
{
g_pDecompressedFileData = new char[MAX_WAVEFILEDATA_LEN];
g_nDecompressedDataBytes = 0;
g_pDecompressedDataFilename = pDecompressedFile;
}
g_bVoiceRecording = false;
if ( g_pVoiceRecord )
{
g_bVoiceRecording = VoiceRecord_Start();
if ( g_bVoiceRecording )
{
if ( steamapicontext && steamapicontext->SteamFriends() && steamapicontext->SteamUser() )
{
// Tell Friends' Voice chat that the local user has started speaking
steamapicontext->SteamFriends()->SetInGameVoiceSpeaking( steamapicontext->SteamUser()->GetSteamID(), true );
}
g_pSoundServices->OnChangeVoiceStatus( -1, true ); // Tell the client DLL.
}
}
return g_bVoiceRecording;
}
void Voice_UserDesiresStop()
{
if ( g_bVoiceRecordStopping )
return;
g_bVoiceRecordStopping = true;
g_pSoundServices->OnChangeVoiceStatus( -1, false ); // Tell the client DLL.
// If we're using Steam voice, we'll keep recording until Steam tells us we
// received all the data.
if ( g_bUsingSteamVoice )
{
steamapicontext->SteamUser()->StopVoiceRecording();
}
else
{
VoiceRecord_Stop();
}
}
bool Voice_RecordStop()
{
// Write the files out for debugging.
if(g_pMicInputFileData)
{
delete [] g_pMicInputFileData;
g_pMicInputFileData = NULL;
}
if(g_pUncompressedFileData)
{
WriteWaveFile(g_pUncompressedDataFilename, g_pUncompressedFileData, g_nUncompressedDataBytes, g_VoiceSampleFormat.wBitsPerSample, g_VoiceSampleFormat.nChannels, Voice_SamplesPerSec() );
delete [] g_pUncompressedFileData;
g_pUncompressedFileData = NULL;
}
if(g_pDecompressedFileData)
{
WriteWaveFile(g_pDecompressedDataFilename, g_pDecompressedFileData, g_nDecompressedDataBytes, g_VoiceSampleFormat.wBitsPerSample, g_VoiceSampleFormat.nChannels, Voice_SamplesPerSec() );
delete [] g_pDecompressedFileData;
g_pDecompressedFileData = NULL;
}
g_VoiceWriter.Finish();
VoiceRecord_Stop();
if ( g_bVoiceRecording )
{
if ( steamapicontext->SteamFriends() && steamapicontext->SteamUser() )
{
// Tell Friends' Voice chat that the local user has stopped speaking
steamapicontext->SteamFriends()->SetInGameVoiceSpeaking( steamapicontext->SteamUser()->GetSteamID(), false );
}
}
g_bVoiceRecording = false;
g_bVoiceRecordStopping = false;
return(true);
}
int Voice_GetCompressedData(char *pchDest, int nCount, bool bFinal)
{
// Check g_bVoiceRecordStopping in case g_bUsingSteamVoice changes on us
// while waiting for the end of voice data.
if ( g_bUsingSteamVoice && g_bVoiceRecordStopping )
{
uint32 cbCompressedWritten = 0;
uint32 cbUncompressedWritten = 0;
uint32 cbCompressed = 0;
uint32 cbUncompressed = 0;
// We're going to always request steam give us the encoded stream at the optimal rate, unless our final output
// rate is lower than it. We'll pass our output rate when we actually extract the data, which Steam will
// happily upsample from its optimal rate for us.
int nEncodeRate = min( (int)steamapicontext->SteamUser()->GetVoiceOptimalSampleRate(), Voice_SamplesPerSec() );
EVoiceResult result = steamapicontext->SteamUser()->GetAvailableVoice( &cbCompressed, &cbUncompressed, nEncodeRate );
if ( result == k_EVoiceResultOK )
{
result = steamapicontext->SteamUser()->GetVoice( true, pchDest, nCount, &cbCompressedWritten,
g_pUncompressedFileData != NULL, g_pUncompressedFileData,
MAX_WAVEFILEDATA_LEN - g_nUncompressedDataBytes,
&cbUncompressedWritten, nEncodeRate );
if ( g_pUncompressedFileData )
{
g_nUncompressedDataBytes += cbUncompressedWritten;
}
g_pSoundServices->OnChangeVoiceStatus( -3, true );
}
else
{
if ( result == k_EVoiceResultNotRecording && g_bVoiceRecording )
{
Voice_RecordStop();
}
g_pSoundServices->OnChangeVoiceStatus( -3, false );
}
return cbCompressedWritten;
}
IVoiceCodec *pCodec = g_pEncodeCodec;
if( g_pVoiceRecord && pCodec )
{
#ifdef VOICE_VOX_ENABLE
static ConVarRef voice_vox( "voice_vox" );
#endif // VOICE_VOX_ENABLE
short tempData[8192];
int samplesWanted = min(nCount/BYTES_PER_SAMPLE, (int)sizeof(tempData)/BYTES_PER_SAMPLE);
int gotten = g_pVoiceRecord->GetRecordedData(tempData, samplesWanted);
// If they want to get the data from a file instead of the mic, use that.
if(g_pMicInputFileData)
{
double curtime = Plat_FloatTime();
int nShouldGet = (curtime - g_MicStartTime) * Voice_SamplesPerSec();
gotten = min(sizeof(tempData)/BYTES_PER_SAMPLE,
(size_t)min(nShouldGet, (g_nMicInputFileBytes - g_CurMicInputFileByte) / BYTES_PER_SAMPLE));
memcpy(tempData, &g_pMicInputFileData[g_CurMicInputFileByte], gotten*BYTES_PER_SAMPLE);
g_CurMicInputFileByte += gotten * BYTES_PER_SAMPLE;
g_MicStartTime = curtime;
}
#ifdef VOICE_VOX_ENABLE
else if ( gotten && voice_vox.GetBool() == true )
{
// If the voice data is essentially silent, don't transmit
short *pData = tempData;
int averageData = 0;
int minData = 16384;
int maxData = -16384;
for ( int i=0; i<gotten; ++i )
{
short val = *pData;
averageData += val;
minData = min( val, minData );
maxData = max( val, maxData );
++pData;
}
averageData /= gotten;
int deltaData = maxData - minData;
if ( deltaData < voice_threshold.GetFloat() && maxData < voice_threshold.GetFloat() )
{
// -3 signals that we're silent
g_pSoundServices->OnChangeVoiceStatus( -3, false );
return 0;
}
}
#endif // VOICE_VOX_ENABLE
#ifdef VOICE_SEND_RAW_TEST
int nCompressedBytes = min( gotten, nCount );
for ( int i=0; i < nCompressedBytes; i++ )
{
pchDest[i] = (char)(tempData[i] >> 8);
}
#else
int nCompressedBytes = pCodec->Compress((char*)tempData, gotten, pchDest, nCount, !!bFinal);
#endif
// Write to our file buffers..
if(g_pUncompressedFileData)
{
int nToWrite = min(gotten*BYTES_PER_SAMPLE, MAX_WAVEFILEDATA_LEN - g_nUncompressedDataBytes);
memcpy(&g_pUncompressedFileData[g_nUncompressedDataBytes], tempData, nToWrite);
g_nUncompressedDataBytes += nToWrite;
}
#ifdef VOICE_VOX_ENABLE
// -3 signals that we're talking
g_pSoundServices->OnChangeVoiceStatus( -3, (nCompressedBytes > 0) );
#endif // VOICE_VOX_ENABLE
return nCompressedBytes;
}
else
{
#ifdef VOICE_VOX_ENABLE
// -3 signals that we're silent
g_pSoundServices->OnChangeVoiceStatus( -3, false );
#endif // VOICE_VOX_ENABLE
return 0;
}
}
//------------------ Copyright (c) 1999 Valve, LLC. ----------------------------
// Purpose: Assigns a channel to an entity by searching for either a channel
// already assigned to that entity or picking the least recently used
// channel. If the LRU channel is picked, it is flushed and all other
// channels are aged.
// Input : nEntity - entity number to assign to a channel.
// Output : A channel index to which the entity has been assigned.
//------------------------------------------------------------------------------
int Voice_AssignChannel(int nEntity, bool bProximity)
{
if(g_bInTweakMode)
return VOICE_CHANNEL_IN_TWEAK_MODE;
// See if a channel already exists for this entity and if so, just return it.
int iFree = -1;
for(int i=0; i < VOICE_NUM_CHANNELS; i++)
{
CVoiceChannel *pChannel = &g_VoiceChannels[i];
if(pChannel->m_iEntity == nEntity)
{
return i;
}
else if(pChannel->m_iEntity == -1 && ( pChannel->m_pVoiceCodec || g_bUsingSteamVoice ) )
{
// Won't exist in steam voice mode
if ( pChannel->m_pVoiceCodec )
{
pChannel->m_pVoiceCodec->ResetState();
}
iFree = i;
break;
}
}
// If they're all used, then don't allow them to make a new channel.
if(iFree == -1)
{
return VOICE_CHANNEL_ERROR;
}
CVoiceChannel *pChannel = &g_VoiceChannels[iFree];
pChannel->Init(nEntity);
pChannel->m_bProximity = bProximity;
VoiceSE_StartOverdrive();
return iFree;
}
//------------------ Copyright (c) 1999 Valve, LLC. ----------------------------
// Purpose: Determines which channel has been assigened to a given entity.
// Input : nEntity - entity number.
// Output : The index of the channel assigned to the entity, VOICE_CHANNEL_ERROR
// if no channel is currently assigned to the given entity.
//------------------------------------------------------------------------------
int Voice_GetChannel(int nEntity)
{
for(int i=0; i < VOICE_NUM_CHANNELS; i++)
if(g_VoiceChannels[i].m_iEntity == nEntity)
return i;
return VOICE_CHANNEL_ERROR;
}
double UpsampleIntoBuffer(
const short *pSrc,
int nSrcSamples,
CCircularBuffer *pBuffer,
double startFraction,
double rate)
{
double maxFraction = nSrcSamples - 1;
while(1)
{
if(startFraction >= maxFraction)
break;
int iSample = (int)startFraction;
double frac = startFraction - floor(startFraction);
double val1 = pSrc[iSample];
double val2 = pSrc[iSample+1];
short newSample = (short)(val1 + (val2 - val1) * frac);
pBuffer->Write(&newSample, sizeof(newSample));
startFraction += rate;
}
return startFraction - floor(startFraction);
}
//------------------ Copyright (c) 1999 Valve, LLC. ----------------------------
// Purpose: Adds received voice data to
// Input :
// Output :
//------------------------------------------------------------------------------
int Voice_AddIncomingData(int nChannel, const char *pchData, int nCount, int iSequenceNumber)
{
CVoiceChannel *pChannel;
// If in tweak mode, we call this during Idle with -1 as the channel, so any channel data from the network
// gets rejected.
if(g_bInTweakMode)
{
if(nChannel == TWEAKMODE_CHANNELINDEX)
nChannel = 0;
else
return 0;
}
if ( ( pChannel = GetVoiceChannel(nChannel)) == NULL || ( !g_bUsingSteamVoice && !pChannel->m_pVoiceCodec ) )
{
return(0);
}
pChannel->m_bStarved = false; // This only really matters if you call Voice_AddIncomingData between the time the mixer
// asks for data and Voice_Idle is called.
// Decompress.
// @note Tom Bui: suggested destination buffer for Steam voice is 22kb
char decompressed[22528];
#ifdef VOICE_SEND_RAW_TEST
int nDecompressed = nCount;
for ( int i=0; i < nDecompressed; i++ )
((short*)decompressed)[i] = pchData[i] << 8;
#else
int nDecompressed = 0;
if ( g_bUsingSteamVoice )
{
uint32 nBytesWritten = 0;
EVoiceResult result = steamapicontext->SteamUser()->DecompressVoice( pchData, nCount,
decompressed, sizeof( decompressed ),
&nBytesWritten, Voice_SamplesPerSec() );
if ( result == k_EVoiceResultOK )
{
nDecompressed = nBytesWritten / BYTES_PER_SAMPLE;
}
}
else
{
nDecompressed = pChannel->m_pVoiceCodec->Decompress(pchData, nCount, decompressed, sizeof(decompressed));
}
#endif
if ( g_bInTweakMode )
{
short *data = (short *)decompressed;
g_VoiceTweakSpeakingVolume = 0;
// Find the highest value
for ( int i=0; i<nDecompressed; ++i )
{
g_VoiceTweakSpeakingVolume = max((int)abs(data[i]), g_VoiceTweakSpeakingVolume);
}
// Smooth it out
g_VoiceTweakSpeakingVolume &= 0xFE00;
}
pChannel->m_AutoGain.ProcessSamples((short*)decompressed, nDecompressed);
// Upsample into the dest buffer. We could do this in a mixer but it complicates the mixer.
pChannel->m_LastFraction = UpsampleIntoBuffer( (short*)decompressed,
nDecompressed,
&pChannel->m_Buffer,
pChannel->m_LastFraction,
(double)Voice_SamplesPerSec()/g_VoiceSampleFormat.nSamplesPerSec );
pChannel->m_LastSample = decompressed[nDecompressed];
// Write to our file buffer..
if(g_pDecompressedFileData)
{
int nToWrite = min(nDecompressed*2, MAX_WAVEFILEDATA_LEN - g_nDecompressedDataBytes);
memcpy(&g_pDecompressedFileData[g_nDecompressedDataBytes], decompressed, nToWrite);
g_nDecompressedDataBytes += nToWrite;
}
g_VoiceWriter.AddDecompressedData( pChannel, (const byte *)decompressed, nDecompressed * 2 );
if( voice_showincoming.GetInt() != 0 )
{
Msg("Voice - %d incoming samples added to channel %d\n", nDecompressed, nChannel);
}
return(nChannel);
}
#if DEAD
//------------------ Copyright (c) 1999 Valve, LLC. ----------------------------
// Purpose: Flushes a given receive channel.
// Input : nChannel - index of channel to flush.
//------------------------------------------------------------------------------
void Voice_FlushChannel(int nChannel)
{
if ((nChannel < 0) || (nChannel >= VOICE_NUM_CHANNELS))
{
Assert(false);
return;
}
g_VoiceChannels[nChannel].m_Buffer.Flush();
}
#endif
//------------------------------------------------------------------------------
// IVoiceTweak implementation.
//------------------------------------------------------------------------------
int VoiceTweak_StartVoiceTweakMode()
{
// If we're already in voice tweak mode, return an error.
if(g_bInTweakMode)
{
Assert(!"VoiceTweak_StartVoiceTweakMode called while already in tweak mode.");
return 0;
}
if ( !g_pMixerControls && voice_enable.GetBool() )
{
Voice_ForceInit();
}
if( !g_pMixerControls )
return 0;
Voice_EndAllChannels();
Voice_RecordStart(NULL, NULL, NULL);
Voice_AssignChannel(TWEAKMODE_ENTITYINDEX, false );
g_bInTweakMode = true;
InitMixerControls();
return 1;
}
void VoiceTweak_EndVoiceTweakMode()
{
if(!g_bInTweakMode)
{
Assert(!"VoiceTweak_EndVoiceTweakMode called when not in tweak mode.");
return;
}
g_bInTweakMode = false;
Voice_RecordStop();
}
void VoiceTweak_SetControlFloat(VoiceTweakControl iControl, float flValue)
{
if(!g_pMixerControls)
return;
if(iControl == MicrophoneVolume)
{
g_pMixerControls->SetValue_Float(IMixerControls::MicVolume, flValue);
}
else if ( iControl == MicBoost )
{
g_pMixerControls->SetValue_Float( IMixerControls::MicBoost, flValue );
}
else if(iControl == OtherSpeakerScale)
{
voice_scale.SetValue( flValue );
}
}
void Voice_ForceInit()
{
if ( g_pMixerControls || !voice_enable.GetBool() )
{
// Nothing to do
return;
}
// Lacking a better default, just peak at what the server's sv_voicecodec is set to
static ConVarRef sv_voicecodec( "sv_voicecodec" );
if ( !Voice_InitWithDefault( sv_voicecodec.GetString() ) )
{
// Try ultimate fallback
Voice_InitWithDefault( VOICE_FALLBACK_CODEC );
}
}
float VoiceTweak_GetControlFloat(VoiceTweakControl iControl)
{
Voice_ForceInit();
if(!g_pMixerControls)
return 0;
if(iControl == MicrophoneVolume)
{
float value = 1;
g_pMixerControls->GetValue_Float(IMixerControls::MicVolume, value);
return value;
}
else if(iControl == OtherSpeakerScale)
{
return voice_scale.GetFloat();
}
else if(iControl == SpeakingVolume)
{
return g_VoiceTweakSpeakingVolume * 1.0f / 32768;
}
else if ( iControl == MicBoost )
{
float flValue = 1;
g_pMixerControls->GetValue_Float( IMixerControls::MicBoost, flValue );
return flValue;
}
else
{
return 1;
}
}
bool VoiceTweak_IsStillTweaking()
{
return g_bInTweakMode;
}
void Voice_Spatialize( channel_t *channel )
{
if ( !g_bInTweakMode )
return;
Assert( channel->sfx );
Assert( channel->sfx->pSource );
Assert( channel->sfx->pSource->GetType() == CAudioSource::AUDIO_SOURCE_VOICE );
// Place the tweak mode sound back at the view entity
CVoiceChannel *pVoiceChannel = GetVoiceChannel( 0 );
Assert( pVoiceChannel );
if ( !pVoiceChannel )
return;
if ( pVoiceChannel->m_nSoundGuid != channel->guid )
return;
// No change
if ( g_pSoundServices->GetViewEntity() == pVoiceChannel->m_nViewEntityIndex )
return;
DevMsg( 1, "Voice_Spatialize changing voice tweak entity from %d to %d\n", pVoiceChannel->m_nViewEntityIndex, g_pSoundServices->GetViewEntity() );
pVoiceChannel->m_nViewEntityIndex = g_pSoundServices->GetViewEntity();
channel->soundsource = pVoiceChannel->m_nViewEntityIndex;
}
IVoiceTweak g_VoiceTweakAPI =
{
VoiceTweak_StartVoiceTweakMode,
VoiceTweak_EndVoiceTweakMode,
VoiceTweak_SetControlFloat,
VoiceTweak_GetControlFloat,
VoiceTweak_IsStillTweaking,
};