Merge branch 'main' into 407-verify-and-possibly-fix-metadata-access-for-rotateaziele_dirac (9619a7d5) · Commits · IVAS Codec Public Collaboration / IVAS Codec

apps/decoder.c

+64 −9

Original line number	Diff line number	Diff line
		@@ -106,7 +106,12 @@ typedef struct
		bool customLsOutputEnabled;
		char *customLsSetupFilename;
		int16_t orientation_tracking;
		#ifdef NON_DIEGETIC_PAN
		int16_t Opt_non_diegetic_pan;
		float non_diegetic_pan_gain;
		#else
		float no_diegetic_pan;
		#endif
		bool renderConfigEnabled;
		char *renderConfigFilename;
		#ifdef COMPLEXITY_LEVEL_INDICATION
		@@ -193,7 +198,11 @@ int main(
		* Open decoder handle
		------------------------------------------------------------------------------------------/

		#ifdef NON_DIEGETIC_PAN
		if ( ( error = IVAS_DEC_Open( &hIvasDec, arg.decMode, arg.orientation_tracking ) ) != IVAS_ERR_OK )
		#else
		if ( ( error = IVAS_DEC_Open( &hIvasDec, arg.decMode, arg.orientation_tracking, arg.no_diegetic_pan ) ) != IVAS_ERR_OK )
		#endif
		{
		fprintf( stderr, "Open failed: %s\n", IVAS_DEC_GetErrorMessage( error ) );
		goto cleanup;
		@@ -358,7 +367,11 @@ int main(
		* Configure the decoder
		------------------------------------------------------------------------------------------/

		#ifdef NON_DIEGETIC_PAN
		if ( ( error = IVAS_DEC_Configure( hIvasDec, arg.output_Fs, arg.outputFormat, arg.customLsOutputEnabled, arg.hrtfReaderEnabled, arg.enableHeadRotation, arg.renderConfigEnabled, arg.Opt_non_diegetic_pan, arg.non_diegetic_pan_gain ) ) != IVAS_ERR_OK )
		#else
		if ( ( error = IVAS_DEC_Configure( hIvasDec, arg.output_Fs, arg.outputFormat, arg.customLsOutputEnabled, arg.hrtfReaderEnabled, arg.enableHeadRotation, arg.renderConfigEnabled ) ) != IVAS_ERR_OK )
		#endif
		{
		fprintf( stderr, "\nConfigure failed: %s\n\n", IVAS_DEC_GetErrorMessage( error ) );
		goto cleanup;
		@@ -793,7 +806,12 @@ static bool parseCmdlIVAS_dec(
		arg->renderConfigFilename = NULL;

		arg->inputFormat = IVAS_DEC_INPUT_FORMAT_G192;
		#ifdef NON_DIEGETIC_PAN
		arg->Opt_non_diegetic_pan = 0;
		arg->non_diegetic_pan_gain = 0.f;
		#else
		arg->no_diegetic_pan = 0.f;
		#endif

		/-----------------------------------------------------------------
		* Initialization
		@@ -1019,38 +1037,61 @@ static bool parseCmdlIVAS_dec(
		}
		i += 2;
		}
		else if ( strcmp( argv_to_upper, "-NO_DIEGETIC_PAN" ) == 0 )
		else if ( strcmp( argv_to_upper, "-NON_DIEGETIC_PAN" ) == 0 )
		{
		i++;

		if ( argc - i <= 4 \|\| argv[i][0] == '-' )
		#ifdef NON_DIEGETIC_PAN
		arg->Opt_non_diegetic_pan = 1;
		#else
		if ( argc - i <= 4 \|\| ( argv[i][0] == '-' ) )
		{
		fprintf( stderr, "Error: Argument for panning option not specified!\n\n" );
		usage_dec();
		return false;
		}

		#endif
		strncpy( argv_to_upper, argv[i], sizeof( argv_to_upper ) - 1 );
		argv_to_upper[sizeof( argv_to_upper ) - 1] = '\0';
		#ifdef NON_DIEGETIC_PAN
		to_upper( argv_to_upper );
		#endif
		if ( ( strcmp( argv_to_upper, "CENTER" ) == 0 ) \|\| ( strchr( argv_to_upper, 'C' ) != NULL ) )
		{
		#ifdef NON_DIEGETIC_PAN
		arg->non_diegetic_pan_gain = 0.f;
		#else
		arg->no_diegetic_pan = 0.f;
		#endif
		}
		else if ( ( strcmp( argv_to_upper, "LEFT" ) == 0 ) \|\| ( strchr( argv_to_upper, 'L' ) != NULL ) )
		{
		arg->no_diegetic_pan = -1.f;
		#ifdef NON_DIEGETIC_PAN
		arg->non_diegetic_pan_gain = 1.f;
		#else
		arg->no_diegetic_pan = 1.f;
		#endif
		}
		else if ( ( strcmp( argv_to_upper, "RIGHT" ) == 0 ) \|\| ( strchr( argv_to_upper, 'R' ) != NULL ) )
		{
		arg->no_diegetic_pan = 1.f;
		#ifdef NON_DIEGETIC_PAN
		arg->non_diegetic_pan_gain = -1.f;
		#else
		arg->no_diegetic_pan = -1.f;
		#endif
		}
		else
		{
		#ifdef NON_DIEGETIC_PAN
		arg->non_diegetic_pan_gain = (float) atof( argv_to_upper ) / 90.f;

		if ( arg->non_diegetic_pan_gain > 1.0f \|\| arg->non_diegetic_pan_gain < -1.0f )
		#else
		arg->no_diegetic_pan = (float) atof( argv_to_upper );

		if ( arg->no_diegetic_pan > 1.0f \|\| arg->no_diegetic_pan < -1.0f )
		#endif
		{
		fprintf( stderr, "Error: Incorrect value for panning option argument specified: %s\n\n", argv[i] );
		fprintf( stderr, "Error: Incorrect value for panning gain value specified: %s\n\n", argv[i] );
		usage_dec();
		return false;
		}
		@@ -1104,11 +1145,25 @@ static bool parseCmdlIVAS_dec(
		arg->customLsSetupFilename = argv[i];
		}
		i++;
		#ifdef NON_DIEGETIC_PAN
		if ( ( arg->Opt_non_diegetic_pan ) && ( arg->outputFormat != IVAS_DEC_OUTPUT_STEREO ) )
		{
		fprintf( stderr, "Error: non-diegetic panning is supported in stereo only\n\n" );
		usage_dec();
		return false;
		}
		#endif
		}
		else
		{
		arg->outputFormat = IVAS_DEC_OUTPUT_MONO;
		arg->decMode = IVAS_DEC_MODE_EVS;
		#ifdef NON_DIEGETIC_PAN
		if ( ( arg->Opt_non_diegetic_pan ) )
		{
		arg->outputFormat = IVAS_DEC_OUTPUT_STEREO;
		}
		#endif
		}

		/-----------------------------------------------------------------
		@@ -1216,8 +1271,8 @@ static void usage_dec( void )
		fprintf( stdout, "-rvf File : Reference vector specified by external trajectory file\n" );
		fprintf( stdout, " works only in combination with '-otr ref_vec' and 'ref_vec_lev' modes\n" );
		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" );
		fprintf( stdout, "-non_diegetic_pan P : panning mono non-diegetic sound to stereo with paning P, -90<= P <=90,\n" );
		fprintf( stdout, " left or l or 90->left, right or r or -90->right, center or c or 0->middle\n" );
		fprintf( stdout, "-q : Quiet mode, no frame counter\n" );
		fprintf( stdout, " default is deactivated\n" );
		#ifdef DEBUGGING

apps/renderer.c

+92 −5

Original line number	Diff line number	Diff line
		@@ -136,7 +136,12 @@ typedef struct
		char customHrtfFilePath[RENDERER_MAX_CLI_ARG_LENGTH];
		char renderConfigFilePath[RENDERER_MAX_CLI_ARG_LENGTH];
		int8_t orientationTracking;
		#ifdef NON_DIEGETIC_PAN
		int16_t nonDiegeticPan;
		float nonDiegeticPanGain;
		#else
		float noDiegeticPan;
		#endif
		bool delayCompensationEnabled;
		bool quietModeEnabled;
		bool sceneDescriptionInput;
		@@ -160,7 +165,7 @@ typedef enum
		CmdLnOptionId_refRotFile,
		CmdLnOptionId_customHrtfFile,
		CmdLnOptionId_renderConfigFile,
		CmdLnOptionId_noDiegeticPan,
		CmdLnOptionId_nonDiegeticPan,
		CmdLnOptionId_orientationTracking,
		CmdlnOptionId_lfePosition,
		CmdlnOptionId_lfeMatrix,
		@@ -234,10 +239,10 @@ static const CmdLnParser_Option cliOptions[] = {
		.description = "Binaural renderer configuration file (only for BINAURAL and BINAURAL_ROOM outputs)",
		},
		{
		.id = CmdLnOptionId_noDiegeticPan,
		.match = "no_diegetic_pan",
		.id = CmdLnOptionId_nonDiegeticPan,
		.match = "non_diegetic_pan",
		.matchShort = "ndp",
		.description = "Panning mono no diegetic sound to stereo -1<= pan <= 1\nleft or l or 1->left, right or r or -1->right, center or c or 0 ->middle\n(todo: implementation)",
		.description = "Panning mono non diegetic sound to stereo -90<= pan <= 90\nleft or l or 90->left, right or r or -90->right, center or c or 0 ->middle\n(todo: implementation)",
		},
		{
		.id = CmdLnOptionId_orientationTracking,
		@@ -551,6 +556,21 @@ int main(

		CmdlnArgs args = parseCmdlnArgs( argc, argv );

		#ifdef NON_DIEGETIC_PAN
		if ( args.nonDiegeticPan && !( ( args.inConfig.numAudioObjects == 0 && args.inConfig.multiChannelBuses[0].audioConfig == IVAS_REND_AUDIO_CONFIG_MONO ) \|\|
		( args.inConfig.numAudioObjects > 0 && args.inConfig.audioObjects[0].audioConfig == IVAS_REND_AUDIO_CONFIG_OBJECT && args.inConfig.numAudioObjects == 1 ) ) )
		{
		fprintf( stderr, "\ninvalid configuration - non-diegetic panning requires mono or ISM1 input\n" );
		exit( -1 );
		}

		if ( args.nonDiegeticPan && args.outConfig.audioConfig != IVAS_REND_AUDIO_CONFIG_STEREO )
		{
		fprintf( stderr, "\ninvalid configuration - non-diegetic panning requires stereo output\n" );
		exit( -1 );
		}
		#endif

		positionProvider = IsmPositionProvider_open();

		convert_backslash( args.inputFilePath );
		@@ -665,7 +685,11 @@ int main(
		IVAS_REND_InputId sbaIds[RENDERER_MAX_SBA_INPUTS] = { 0 };
		IVAS_REND_InputId masaIds[RENDERER_MAX_MASA_INPUTS] = { 0 };

		#ifdef NON_DIEGETIC_PAN
		if ( ( error = IVAS_REND_Open( &hIvasRend, args.sampleRate, args.outConfig.audioConfig, args.nonDiegeticPan, args.nonDiegeticPanGain ) ) != IVAS_ERR_OK )
		#else
		if ( ( error = IVAS_REND_Open( &hIvasRend, args.sampleRate, args.outConfig.audioConfig ) ) != IVAS_ERR_OK )
		#endif
		{
		fprintf( stderr, "Error opening renderer handle: %s\n", ivas_error_to_string( error ) );
		exit( -1 );
		@@ -1334,6 +1358,38 @@ static bool parseOutConfig(
		return true;
		}

		#ifdef NON_DIEGETIC_PAN
		static bool parseDiegeticPan(
		char *value,
		float *nonDiegeticPan )
		{
		to_upper( value );

		if ( ( strcmp( value, "CENTER" ) == 0 ) \|\| ( strchr( value, 'C' ) != NULL ) )
		{
		*nonDiegeticPan = 0.f;
		}
		else if ( ( strcmp( value, "LEFT" ) == 0 ) \|\| ( strchr( value, 'L' ) != NULL ) )
		{
		*nonDiegeticPan = 1.f;
		}
		else if ( ( strcmp( value, "RIGHT" ) == 0 ) \|\| ( strchr( value, 'R' ) != NULL ) )
		{
		*nonDiegeticPan = -1.f;
		}
		else
		{
		*nonDiegeticPan = (float) atof( value ) / 90.f;

		if ( nonDiegeticPan > 1.0f \|\| nonDiegeticPan < -1.0f )
		{
		fprintf( stderr, "Error: Incorrect value for panning option argument specified!\n\n" );
		return false;
		}
		}
		return true;

		#else
		static bool parseDiegeticPan(
		char *value,
		float *noDiegeticPan )
		@@ -1363,6 +1419,7 @@ static bool parseDiegeticPan(
		}
		}
		return false;
		#endif
		}

		static bool parseOrientationTracking(
		@@ -1615,7 +1672,12 @@ static CmdlnArgs defaultArgs(
		clearString( args.renderConfigFilePath );

		args.orientationTracking = IVAS_ORIENT_TRK_NONE;
		#ifdef NON_DIEGETIC_PAN
		args.nonDiegeticPan = 0;
		args.nonDiegeticPanGain = 0.f;
		#else
		args.noDiegeticPan = 0.0f;
		#endif

		args.delayCompensationEnabled = true;
		args.quietModeEnabled = false;
		@@ -1705,13 +1767,20 @@ static void parseOption(
		assert( numOptionValues == 1 );
		strncpy( args->renderConfigFilePath, optionValues[0], RENDERER_MAX_CLI_ARG_LENGTH - 1 );
		break;
		case CmdLnOptionId_noDiegeticPan:
		case CmdLnOptionId_nonDiegeticPan:
		assert( numOptionValues == 1 );
		#ifdef NON_DIEGETIC_PAN
		if ( !parseDiegeticPan( optionValues[0], &args->nonDiegeticPanGain ) )
		#else
		if ( !parseDiegeticPan( optionValues[0], &args->noDiegeticPan ) )
		#endif
		{
		fprintf( stderr, "Unknown option for diegetic panning: %s\n", optionValues[0] );
		exit( -1 );
		}
		#ifdef NON_DIEGETIC_PAN
		args->nonDiegeticPan = 1;
		#endif
		break;
		case CmdLnOptionId_orientationTracking:
		assert( numOptionValues == 1 );
		@@ -1822,6 +1891,9 @@ void getMetadataFromFileReader(
		objectMetadataBuffer->positions[objIdx].radius = ismMetadata.radius;
		objectMetadataBuffer->positions[objIdx].yaw = ismMetadata.yaw;
		objectMetadataBuffer->positions[objIdx].pitch = ismMetadata.pitch;
		#ifdef ISM_NON_DIEGETIC_PAN
		objectMetadataBuffer->positions[objIdx].non_diegetic_flag = ismMetadata.non_diegetic_flag;
		#endif

		return;
		}
		@@ -1879,6 +1951,9 @@ static void IsmPositionProvider_getNextFrame(
		objectMetadataBuffer->positions[objIdx].radius = 1.0f;
		objectMetadataBuffer->positions[objIdx].yaw = 0.0f;
		objectMetadataBuffer->positions[objIdx].pitch = 0.0f;
		#ifdef ISM_NON_DIEGETIC_PAN
		objectMetadataBuffer->positions[objIdx].non_diegetic_flag = 0;
		#endif
		}

		/* Wrap azimuth to lie within (-180, 180] range */
		@@ -2140,13 +2215,22 @@ static void parseObjectPosition(
		{
		char *endptr;
		int16_t read_values;
		#ifdef ISM_NON_DIEGETIC_PAN
		float meta_prm[8] = { 0.0f, 0.0f, 1.0f, 0.0f, 1.0f, 0.0f, 0.0f, 0.0f };
		#else
		float meta_prm[7] = { 0.0f, 0.0f, 1.0f, 0.0f, 1.0f, 0.0f, 0.0f };
		#endif


		readNextMetadataChunk( line, "," );
		*positionDuration = (uint16_t) strtol( line, &endptr, 10 );
		readNextMetadataChunk( line, "\n" );

		#ifdef ISM_NON_DIEGETIC_PAN
		read_values = (int16_t) sscanf( line, "%f,%f,%f,%f,%f,%f,%f,%f", &meta_prm[0], &meta_prm[1], &meta_prm[2], &meta_prm[3], &meta_prm[4], &meta_prm[5], &meta_prm[6], &meta_prm[7] );
		#else
		read_values = (int16_t) sscanf( line, "%f,%f,%f,%f,%f,%f,%f", &meta_prm[0], &meta_prm[1], &meta_prm[2], &meta_prm[3], &meta_prm[4], &meta_prm[5], &meta_prm[6] );
		#endif

		if ( read_values < 2 )
		{
		@@ -2159,6 +2243,9 @@ static void parseObjectPosition(
		position->radius = meta_prm[2];
		position->yaw = meta_prm[5];
		position->pitch = meta_prm[6];
		#ifdef ISM_NON_DIEGETIC_PAN
		position->non_diegetic_flag = (int16_t) meta_prm[7];
		#endif
		return;
		}

lib_com/common_api_types.h

+6 −0

Original line number	Diff line number	Diff line
		@@ -80,6 +80,9 @@ typedef struct _IVAS_ISM_METADATA
		float gainFactor;
		float yaw;
		float pitch;
		#ifdef ISM_NON_DIEGETIC_PAN
		int16_t non_diegetic_flag;
		#endif
		} IVAS_ISM_METADATA;

		typedef struct
		@@ -118,6 +121,9 @@ typedef struct
		float radius;
		float yaw;
		float pitch;
		#ifdef ISM_NON_DIEGETIC_PAN
		int16_t non_diegetic_flag;
		#endif
		} IVAS_REND_AudioObjectPosition;

		typedef struct _IVAS_ROOM_ACOUSTICS_CONFIG

lib_com/hq_tools.c

+0 −1

Original line number	Diff line number	Diff line
		@@ -1290,7 +1290,6 @@ int16_t calc_nor_delta_hf(
		ynrm[i] += delta;
		add_bits_denv += bitsforDelta;


		temp_num++;
		}
		}

lib_com/ivas_cnst.h

+22 −0

Original line number	Diff line number	Diff line
		@@ -156,7 +156,12 @@ typedef enum
		RENDERER_MCMASA_MONO_STEREO,
		RENDERER_PARAM_ISM,
		RENDERER_BINAURAL_MIXER_CONV,
		#if defined NON_DIEGETIC_PAN \|\| defined ISM_NON_DIEGETIC_PAN
		RENDERER_BINAURAL_MIXER_CONV_ROOM,
		RENDERER_NON_DIEGETIC_DOWNMIX
		#else
		RENDERER_BINAURAL_MIXER_CONV_ROOM
		#endif

		} RENDERER_TYPE;

		@@ -170,6 +175,9 @@ typedef enum
		#define HEAD_ROTATION_HOA_ORDER 3 /* HOA 3rd order */
		#define MAX_CICP_CHANNELS 16 /* max channels for loudspeaker layouts (16 for custom layouts)*/
		#define MAX_OUTPUT_CHANNELS 16 /* Maximum number of output channels (HOA 3rd order) */
		#ifdef NON_DIEGETIC_PAN
		#define MAX_OUTPUT_CHANNELS_IN_DIEGETIC_PAN 2 /* Maximum number of output channels with non diegetic panning */
		#endif

		#define BINAURAL_CHANNELS 2 /* number of channels for binaural output configuration */
		#define CPE_CHANNELS 2 /* number of CPE (stereo) channels */
		@@ -332,6 +340,9 @@ typedef enum
		#define ISM_RADIUS_MIN 0.0f
		#define ISM_RADIUS_DELTA 0.25f /* Max radius = (2^ISM_RADIUS_NBITS-1)0.25 = 15.75 /
		#define ISM_EXTENDED_METADATA_BRATE IVAS_64k
		#ifdef ISM_NON_DIEGETIC_PAN
		#define ISM_METADATA_IS_NDP_BITS 1
		#endif
		#define ISM_EXTENDED_METADATA_BITS 1
		#define ISM_METADATA_RS_MAX_FRAMES 5 /* Number of frames with opposite extended metadata flags before switching */

		@@ -373,6 +384,9 @@ enum
		{
		IND_ISM_NUM_OBJECTS,
		IND_ISM_EXTENDED_FLAG = IND_ISM_NUM_OBJECTS + MAX_NUM_OBJECTS,
		#ifdef ISM_NON_DIEGETIC_PAN
		IND_ISM_EXTENDED_NDP_FLAG,
		#endif
		IND_ISM_METADATA_FLAG,
		IND_ISM_VAD_FLAG = IND_ISM_METADATA_FLAG + MAX_NUM_OBJECTS,
		IND_ISM_NOISY_SPEECH_FLAG = IND_ISM_VAD_FLAG + MAX_NUM_OBJECTS,
		@@ -381,6 +395,9 @@ enum

		/* ------------- loop for objects -------------- */
		TAG_ISM_LOOP_START = IND_ISM_DTX_COH_SCA + MAX_NUM_OBJECTS,
		#ifdef ISM_NON_DIEGETIC_PAN
		IND_ISM_NDP_FLAG = TAG_ISM_LOOP_START,
		#endif
		IND_ISM_AZIMUTH_DIFF_FLAG = TAG_ISM_LOOP_START,
		IND_ISM_AZIMUTH = TAG_ISM_LOOP_START,
		IND_ISM_ELEVATION_DIFF_FLAG = TAG_ISM_LOOP_START,
		@@ -1572,7 +1589,12 @@ typedef enum
		typedef enum
		{
		TDREND_PLAYSTATUS_INITIAL,
		#ifdef ISM_NON_DIEGETIC_PAN
		TDREND_PLAYSTATUS_PLAYING,
		TDREND_PLAYSTATUS_PLAYING_NON_DIEGETIC
		#else
		TDREND_PLAYSTATUS_PLAYING
		#endif
		} TDREND_PlayStatus_t;

		typedef enum