adding PCM output support to ivas SBA decoder in split rend mode (03186a11) · Commits · IVAS Codec Public Collaboration / IVAS Codec

apps/decoder.c

+63 −8

Original line number	Diff line number	Diff line
		@@ -117,7 +117,9 @@ typedef struct
		float no_diegetic_pan;
		bool renderConfigEnabled;
		char *renderConfigFilename;

		#ifdef SPLIT_REND_WITH_HEAD_ROT
		char *outputMdFilename;
		#endif
		#ifdef DEBUGGING
		IVAS_DEC_FORCED_REND_MODE forcedRendMode;
		#ifdef DEBUG_FOA_AGC
		@@ -267,7 +269,7 @@ int main(
		/* sanity check */
		if ( arg.outputFormat != IVAS_DEC_OUTPUT_BINAURAL && arg.outputFormat != IVAS_DEC_OUTPUT_BINAURAL_ROOM
		#ifdef SPLIT_REND_WITH_HEAD_ROT
		&& arg.outputFormat != IVAS_DEC_OUTPUT_SPLIT_BINAURAL_CLDFB
		&& arg.outputFormat != IVAS_DEC_OUTPUT_SPLIT_BINAURAL_CLDFB && arg.outputFormat != IVAS_DEC_OUTPUT_SPLIT_BINAURAL_PCM
		#endif
		)
		{
		@@ -358,7 +360,7 @@ int main(
		/* sanity check */
		if ( arg.outputFormat != IVAS_DEC_OUTPUT_BINAURAL && arg.outputFormat != IVAS_DEC_OUTPUT_BINAURAL_ROOM
		#ifdef SPLIT_REND_WITH_HEAD_ROT
		&& arg.outputFormat != IVAS_DEC_OUTPUT_SPLIT_BINAURAL_CLDFB
		&& arg.outputFormat != IVAS_DEC_OUTPUT_SPLIT_BINAURAL_CLDFB && arg.outputFormat != IVAS_DEC_OUTPUT_SPLIT_BINAURAL_PCM
		#endif
		)
		{
		@@ -377,7 +379,7 @@ int main(
		* Configure the decoder
		------------------------------------------------------------------------------------------/
		#ifdef SPLIT_REND_WITH_HEAD_ROT
		if ( arg.outputFormat == IVAS_DEC_OUTPUT_SPLIT_BINAURAL_CLDFB )
		if ( arg.outputFormat == IVAS_DEC_OUTPUT_SPLIT_BINAURAL_CLDFB \|\| arg.outputFormat == IVAS_DEC_OUTPUT_SPLIT_BINAURAL_PCM )
		{
		arg.enableHeadRotation = true;
		}
		@@ -489,7 +491,10 @@ int main(

		/* sanity check */
		#ifdef SPLIT_REND_WITH_HEAD_ROT
		if ( ( arg.outputFormat != IVAS_DEC_OUTPUT_BINAURAL_ROOM ) && ( arg.outputFormat != IVAS_DEC_OUTPUT_BINAURAL ) && ( arg.outputFormat != IVAS_DEC_OUTPUT_SPLIT_BINAURAL_CLDFB ) )
		if ( ( arg.outputFormat != IVAS_DEC_OUTPUT_BINAURAL_ROOM ) &&
		( arg.outputFormat != IVAS_DEC_OUTPUT_BINAURAL ) &&
		( arg.outputFormat != IVAS_DEC_OUTPUT_SPLIT_BINAURAL_CLDFB ) &&
		( arg.outputFormat != IVAS_DEC_OUTPUT_SPLIT_BINAURAL_PCM ) )
		{
		fprintf( stderr, "\nExternal Renderer Config is supported only when BINAURAL_ROOM is used as output OR when Split rendering mode is enabled. Exiting. \n" );
		exit( -1 );
		@@ -757,10 +762,14 @@ static IVAS_DEC_AUDIO_CONFIG cmdline2config(
		output_config = IVAS_DEC_OUTPUT_BINAURAL;
		}
		#ifdef SPLIT_REND_WITH_HEAD_ROT
		else if ( strcmp( argv_to_upper, "SPLIT_BINAURAL" ) == 0 )
		else if ( strcmp( argv_to_upper, "BINAURAL_SPLIT_CLDFB" ) == 0 )
		{
		output_config = IVAS_DEC_OUTPUT_SPLIT_BINAURAL_CLDFB;
		}
		else if ( strcmp( argv_to_upper, "BINAURAL_SPLIT_PCM" ) == 0 )
		{
		output_config = IVAS_DEC_OUTPUT_SPLIT_BINAURAL_PCM;
		}
		#endif
		else if ( strcmp( argv_to_upper, "BINAURAL_ROOM" ) == 0 )
		{
		@@ -830,6 +839,10 @@ static bool parseCmdlIVAS_dec(
		arg->renderConfigEnabled = false;
		arg->renderConfigFilename = NULL;

		#ifdef SPLIT_REND_WITH_HEAD_ROT
		arg->outputMdFilename = NULL;
		#endif

		arg->inputFormat = IVAS_DEC_INPUT_FORMAT_G192;
		arg->no_diegetic_pan = 0.f;

		@@ -1057,6 +1070,19 @@ static bool parseCmdlIVAS_dec(
		}
		i += 2;
		}
		#ifdef SPLIT_REND_WITH_HEAD_ROT
		else if ( strcmp( argv_to_upper, "-OM" ) == 0 )
		{
		arg->outputMdFilename = argv[i + 1];
		if ( arg->outputMdFilename[0] == '\0' )
		{
		fprintf( stderr, "Error: output metadata file path not specified\n\n" );
		usage_dec();
		return false;
		}
		i += 2;
		}
		#endif
		else if ( strcmp( argv_to_upper, "-NO_DIEGETIC_PAN" ) == 0 )
		{
		i++;
		@@ -1197,7 +1223,7 @@ static void usage_dec( void )
		fprintf( stdout, "---------------------\n" );
		#ifdef SPLIT_REND_WITH_HEAD_ROT
		fprintf( stdout, "OutputConf : Output configuration: MONO, STEREO, 5_1, 7_1, 5_1_2, 5_1_4, 7_1_4, FOA,\n" );
		fprintf( stdout, " HOA2, HOA3, BINAURAL, BINAURAL_ROOM, SPLIT_BINAURAL, EXT\n" );
		fprintf( stdout, " HOA2, HOA3, BINAURAL, BINAURAL_ROOM, BINAURAL_SPLIT_CLDFB, BINAURAL_SPLIT_PCM, EXT\n" );
		#else
		fprintf( stdout, "OutputConf : Output configuration: MONO, STEREO, 5_1, 7_1, 5_1_2, 5_1_4, 7_1_4, FOA,\n" );
		fprintf( stdout, " HOA2, HOA3, BINAURAL, BINAURAL_ROOM, EXT\n" );
		@@ -1242,6 +1268,9 @@ static void usage_dec( void )
		fprintf( stdout, "-render_config File : Renderer configuration File\n" );
		fprintf( stdout, "-no_diegetic_pan : panning mono non-diegetic sound to stereo -1<= pan <=1,\n" );
		fprintf( stdout, " left or l or 1->left, right or r or -1->right, center or c or 0->middle\n" );
		#ifdef SPLIT_REND_WITH_HEAD_ROT
		fprintf( stdout, "-om File : MD output file for BINAURAL_SPLIT_PCM output format\n" );
		#endif
		fprintf( stdout, "-q : Quiet mode, no frame counter\n" );
		fprintf( stdout, " default is deactivated\n" );
		#ifdef DEBUGGING
		@@ -1343,6 +1372,22 @@ static ivas_error initOnFirstGoodFrame(
		else
		#endif
		{
		#ifdef SPLIT_REND_WITH_HEAD_ROT
		if ( arg.outputFormat == IVAS_DEC_OUTPUT_SPLIT_BINAURAL_PCM )
		{
		int16_t delayNumSamples_temp[3];
		int32_t delayTimeScale_temp;
		IVAS_DEC_GetDelay( hIvasDec, delayNumSamples_temp, &delayTimeScale_temp );
		assert( arg.outputMdFilename != NULL );
		error = split_rend_writer_open( hSplitRendFileReadWrite, arg.outputMdFilename, delayNumSamples_temp[0], delayTimeScale_temp );
		if ( error != IVAS_ERR_OK )
		{
		fprintf( stderr, "Could not open split rend metadata file %s\n", arg.outputWavFilename );
		exit( -1 );
		}
		}
		#endif

		/* Open audio writer and write all previously skipped bad frames now that frame size is known */
		if ( ( error = AudioFileWriter_open( ppAfWriter, arg.outputWavFilename, arg.output_Fs, *pNumOutChannels ) ) != IVAS_ERR_OK )
		{
		@@ -1717,7 +1762,7 @@ static ivas_error decodeG192(
		if ( decodedGoodFrame )
		{
		#ifdef SPLIT_REND_WITH_HEAD_ROT
		if ( hSplitRendFileReadWrite != NULL )
		if ( ( hSplitRendFileReadWrite != NULL ) && ( arg.outputFormat == IVAS_DEC_OUTPUT_SPLIT_BINAURAL_CLDFB ) )
		{
		if ( split_rend_write_bitstream_to_file( hSplitRendFileReadWrite, splitRendBits.bits_buf, &splitRendBits.bits_read, &splitRendBits.bits_written ) != IVAS_ERR_OK )
		{
		@@ -1728,6 +1773,16 @@ static ivas_error decodeG192(
		else
		#endif
		{
		#ifdef SPLIT_REND_WITH_HEAD_ROT
		if ( ( hSplitRendFileReadWrite != NULL ) && ( arg.outputFormat == IVAS_DEC_OUTPUT_SPLIT_BINAURAL_PCM ) )
		{
		if ( split_rend_write_bitstream_to_file( hSplitRendFileReadWrite, splitRendBits.bits_buf, &splitRendBits.bits_read, &splitRendBits.bits_written ) != IVAS_ERR_OK )
		{
		fprintf( stderr, "\nUnable to write to bitstream file!\n" );
		exit( -1 );
		}
		}
		#endif
		if ( delayNumSamples < nOutSamples )
		{
		if ( ( error = AudioFileWriter_write( afWriter, &pcmBuf[delayNumSamples * nOutChannels], nOutSamples * nOutChannels - ( delayNumSamples * nOutChannels ) ) ) != IVAS_ERR_OK )

lib_dec/ivas_binRenderer_internal.c

+3 −2

Original line number	Diff line number	Diff line
		@@ -685,7 +685,7 @@ ivas_error ivas_rend_openCldfbRend(
		hBinRenderer->hInputSetup->is_loudspeaker_setup = 0;
		getAudioConfigNumChannels( inConfig, &hBinRenderer->hInputSetup->nchan_out_woLFE );

		if ( out_config == AUDIO_CONFIG_BINAURAL_SPLIT_CLDFB )
		if ( ( out_config == AUDIO_CONFIG_BINAURAL_SPLIT_CLDFB ) \|\| ( out_config == AUDIO_CONFIG_BINAURAL_SPLIT_PCM ) )
		{
		#ifdef SPLIT_REND_WITH_HEAD_ROT_DEBUG
		hBinRenderer->numPoses = pMultiBinPoseData->num_poses + 1;
		@@ -751,7 +751,8 @@ ivas_error ivas_binRenderer_open(
		}

		#ifdef SPLIT_REND_WITH_HEAD_ROT
		if ( st_ivas->hDecoderConfig->output_config == AUDIO_CONFIG_BINAURAL_SPLIT_CLDFB )
		if ( ( st_ivas->hDecoderConfig->output_config == AUDIO_CONFIG_BINAURAL_SPLIT_CLDFB ) \|\|
		( st_ivas->hDecoderConfig->output_config == AUDIO_CONFIG_BINAURAL_SPLIT_PCM ) )
		{
		#ifdef SPLIT_REND_WITH_HEAD_ROT_DEBUG
		hBinRenderer->numPoses = st_ivas->splitBinRend.splitrend.multiBinPoseData.num_poses + 1;

lib_dec/ivas_dec.c

+7 −3

Original line number	Diff line number	Diff line
		@@ -102,7 +102,8 @@ ivas_error ivas_dec(
		output_frame = (int16_t) ( output_Fs / FRAMES_PER_SEC );

		#ifdef SPLIT_REND_WITH_HEAD_ROT
		if ( output_config == AUDIO_CONFIG_BINAURAL_SPLIT_CLDFB )
		if ( ( output_config == AUDIO_CONFIG_BINAURAL_SPLIT_CLDFB ) \|\|
		( output_config == AUDIO_CONFIG_BINAURAL_SPLIT_PCM ) )
		{
		assert( st_ivas->ivas_format == SBA_FORMAT && ( output_Fs == 48000 ) && "split binaural mode is currently supported with SBA format and 48k sampling rate only" );
		ivas_set_split_rend_setup( &st_ivas->splitBinRend, &st_ivas->hRenderConfig->split_rend_config, st_ivas->hHeadTrackData, hSplitRendBits );
		@@ -627,19 +628,22 @@ ivas_error ivas_dec(

		#ifdef SPLIT_REND_WITH_HEAD_ROT
		/split rendering process calls/
		if ( output_config == AUDIO_CONFIG_BINAURAL_SPLIT_CLDFB )
		if ( ( output_config == AUDIO_CONFIG_BINAURAL_SPLIT_CLDFB ) \|\|
		( output_config == AUDIO_CONFIG_BINAURAL_SPLIT_PCM ) )
		{
		IVAS_DEC_SPLIT_REND_WRAPPER *hSplitBinRend;
		int16_t max_band;
		int16_t pcm_out;
		hSplitBinRend = &st_ivas->splitBinRend;
		max_band = (int16_t) ( ( BINAURAL_MAXBANDS * output_Fs ) / 48000 );
		pcm_out = ( output_config == AUDIO_CONFIG_BINAURAL_SPLIT_PCM ) ? 1 : 0;
		ivas_renderMultiBinToSplitBinaural( &hSplitBinRend->splitrend,
		st_ivas->hHeadTrackData->Quaternions,
		st_ivas->hRenderConfig->split_rend_config.splitRendBitRate,
		hSplitBinRend->hSplitRendBits,
		hSplitBinRend->hMultiBinCldfbData->Cldfb_RealBuffer_Binaural,
		hSplitBinRend->hMultiBinCldfbData->Cldfb_ImagBuffer_Binaural,
		max_band, output, 1, 0 );
		max_band, output, 1, pcm_out );

		free( st_ivas->splitBinRend.hMultiBinCldfbData );
		}

lib_dec/ivas_dirac_dec.c

+2 −1

Original line number	Diff line number	Diff line
		@@ -2570,7 +2570,8 @@ void ivas_dirac_dec(
		Cldfb_RealBuffer,
		Cldfb_ImagBuffer );
		#ifdef SPLIT_REND_WITH_HEAD_ROT
		if ( st_ivas->hDecoderConfig->output_config == AUDIO_CONFIG_BINAURAL_SPLIT_CLDFB )
		if ( ( st_ivas->hDecoderConfig->output_config == AUDIO_CONFIG_BINAURAL_SPLIT_CLDFB ) \|\|
		( st_ivas->hDecoderConfig->output_config == AUDIO_CONFIG_BINAURAL_SPLIT_PCM ) )
		{
		int16_t pos_idx;
		#ifdef SPLIT_REND_WITH_HEAD_ROT_DEBUG

lib_dec/ivas_init_dec.c

+7 −5

Original line number	Diff line number	Diff line
		@@ -630,7 +630,7 @@ ivas_error ivas_init_decoder_front(

		if ( st_ivas->hDecoderConfig->output_config == AUDIO_CONFIG_BINAURAL \|\| st_ivas->hDecoderConfig->output_config == AUDIO_CONFIG_BINAURAL_ROOM
		#ifdef SPLIT_REND_WITH_HEAD_ROT
		\|\| st_ivas->hDecoderConfig->output_config == AUDIO_CONFIG_BINAURAL_SPLIT_CLDFB
		\|\| st_ivas->hDecoderConfig->output_config == AUDIO_CONFIG_BINAURAL_SPLIT_CLDFB \|\| st_ivas->hDecoderConfig->output_config == AUDIO_CONFIG_BINAURAL_SPLIT_PCM
		#endif
		)
		{
		@@ -1217,7 +1217,8 @@ ivas_error ivas_init_decoder(
		if ( st_ivas->renderer_type == RENDERER_BINAURAL_FASTCONV \|\| st_ivas->renderer_type == RENDERER_BINAURAL_FASTCONV_ROOM )
		{
		#ifdef SPLIT_REND_WITH_HEAD_ROT
		if ( st_ivas->hDecoderConfig->output_config == AUDIO_CONFIG_BINAURAL_SPLIT_CLDFB )
		if ( ( st_ivas->hDecoderConfig->output_config == AUDIO_CONFIG_BINAURAL_SPLIT_CLDFB ) \|\|
		( st_ivas->hDecoderConfig->output_config == AUDIO_CONFIG_BINAURAL_SPLIT_PCM ) )
		{
		ivas_renderSplitGetMultiBinPoseData(
		&st_ivas->hRenderConfig->split_rend_config,
		@@ -1230,11 +1231,12 @@ ivas_error ivas_init_decoder(
		return error;
		}
		#ifdef SPLIT_REND_WITH_HEAD_ROT
		if ( st_ivas->hDecoderConfig->output_config == AUDIO_CONFIG_BINAURAL_SPLIT_CLDFB )
		if ( ( st_ivas->hDecoderConfig->output_config == AUDIO_CONFIG_BINAURAL_SPLIT_CLDFB ) \|\|
		( st_ivas->hDecoderConfig->output_config == AUDIO_CONFIG_BINAURAL_SPLIT_PCM ) )
		{
		ivas_split_renderer_open( &st_ivas->splitBinRend.splitrend,
		&st_ivas->hRenderConfig->split_rend_config,
		hDecoderConfig->output_Fs, 1, 0 );
		hDecoderConfig->output_Fs, 1, ( st_ivas->hDecoderConfig->output_config == AUDIO_CONFIG_BINAURAL_SPLIT_PCM ) ? 1 : 0 );
		}
		#endif
		}
		@@ -2067,7 +2069,7 @@ static ivas_error doSanityChecks_IVAS(
		if ( !( output_config == AUDIO_CONFIG_BINAURAL \|\|
		output_config == AUDIO_CONFIG_BINAURAL_ROOM
		#ifdef SPLIT_REND_WITH_HEAD_ROT
		\|\| output_config == AUDIO_CONFIG_BINAURAL_SPLIT_CLDFB
		\|\| output_config == AUDIO_CONFIG_BINAURAL_SPLIT_CLDFB \|\| output_config == AUDIO_CONFIG_BINAURAL_SPLIT_PCM
		#endif
		) )
		{