Merge branch 'Framework-maintenance-20220902' into 'main' (2fbd5586) · Commits · IVAS Codec Public Collaboration / IVAS Codec

apps/encoder.c

+1 −1

Original line number	Diff line number	Diff line
		@@ -1618,7 +1618,7 @@ static void usage_enc( void )
		fprintf( stdout, " where 0 = adaptive, 3-100 = fixed in number of frames,\n" );
		fprintf( stdout, " default is deactivated\n" );
		fprintf( stdout, "-dtx : Activate DTX mode with a SID update rate of 8 frames\n" );
		fprintf( stdout, " Note: DTX is currently supported in EVS, DFT/TD stereo, 1 ISm, \n" );
		fprintf( stdout, " Note: DTX is currently supported in EVS, stereo, 1 ISm, \n" );
		fprintf( stdout, " SBA (up to 128kbps) and MASA (up to 128kbps)\n" );
		fprintf( stdout, "-rf p o : Activate channel-aware mode for WB and SWB signal at 13.2kbps, \n" );
		fprintf( stdout, " where FEC indicator, p: LO or HI, and FEC offset, o: 2, 3, 5, or 7 in number of frames.\n" );

lib_com/bitstream.c

+5 −5

Original line number	Diff line number	Diff line
		@@ -2063,7 +2063,7 @@ ivas_error read_indices(
		}
		else if ( k == SIZE_IVAS_BRATE_TBL )
		{
		/temp change for spar DTX/
		/temp change for SPAR DTX/
		if ( total_brate == IVAS_SID_5k )
		{
		st_ivas->element_mode_init = -1;
		@@ -2116,7 +2116,7 @@ ivas_error read_indices(
		}
		else
		{
		sid_upd_bad = 1; /* this frame type may happen in ETSI/3GPP CS cases , a corrupt sid frames */
		sid_upd_bad = 1; /* this frame type may happen in ETSI/3GPP CS cases, a corrupt SID frames */
		}
		}

		@@ -2201,7 +2201,7 @@ ivas_error read_indices(
		/* total_brate= 0 */
		}

		/* handle bad/lost speech frame(and CS bad sid frame) in the decoders CNG synthesis settings pair (total_brate, bfi) */
		/* handle bad/lost speech frame(and CS bad SID frame) in the decoders CNG synthesis settings pair (total_brate, bfi) */
		if ( ( ( CNG != 0 ) && ( ( speech_bad != 0 ) \|\| ( speech_lost != 0 ) ) ) \|\| / SP_BAD or SPEECH_LOST) --> stay in CNG */
		( sid_upd_bad != 0 ) ) /* SID_UPD_BAD --> start CNG */
		{
		@@ -2320,7 +2320,7 @@ static Word32 read_indices_mime_handle_dtx(
		{
		if ( st->bfi )
		{
		sid_upd_bad = 1; /* corrupt sid_first, signaled as bad sid */
		sid_upd_bad = 1; /* corrupt sid_first, signaled as bad SID */
		}
		else
		{
		@@ -2378,7 +2378,7 @@ static Word32 read_indices_mime_handle_dtx(
		}

		/* in CNG */
		/* handle bad speech frame(and bad sid frame) in the decoders CNG synthesis settings pair (total_brate, bfi) */
		/* handle bad speech frame(and bad SID frame) in the decoders CNG synthesis settings pair (total_brate, bfi) */
		if ( ( CNG != 0 && ( speech_bad \|\| speech_lost \|\| no_data ) ) \|\| / SP_BAD or SPEECH_LOST) --> stay in CNG */
		sid_upd_bad ) /* SID_UPD_BAD --> start/stay CNG */
		{

lib_com/ivas_fb_mixer.c

+2 −2

Original line number	Diff line number	Diff line
		@@ -1098,8 +1098,8 @@ static ivas_error ivas_filterbank_setup(

		for ( j = 0; j < IVAS_MAX_NUM_FB_BANDS; j++ )
		{
		pFb->fb_bin_to_band.p_short_stride_num_bins_per_band[j] = 0; /* aka num_active_bins per spar band */
		pFb->fb_bin_to_band.p_short_stride_start_bin_per_band[j] = 0; /* first considered bin index per spar band */
		pFb->fb_bin_to_band.p_short_stride_num_bins_per_band[j] = 0; /* aka num_active_bins per SPAR band */
		pFb->fb_bin_to_band.p_short_stride_start_bin_per_band[j] = 0; /* first considered bin index per SPAR band */
		pFb->fb_bin_to_band.pp_short_stride_bin_to_band[j] = NULL;
		for ( i = 0; i < CLDFB_NO_CHANNELS_MAX; i++ )
		{

lib_com/ivas_prot.h

+24 −42

Original line number	Diff line number	Diff line
		@@ -794,8 +794,8 @@ void ivas_param_ism_enc(
		);

		void ivas_param_ism_enc_close(
		DIRAC_ENC_HANDLE hDirAC /* i/o: encoder DirAC handle */
		,const int32_t input_Fs /* i : input sampling_rate */
		DIRAC_ENC_HANDLE hDirAC, /* i/o: encoder DirAC handle */
		const int32_t input_Fs /* i : input sampling_rate */
		);

		void ivas_param_ism_stereo_dmx(
		@@ -2155,7 +2155,7 @@ void stereo_mdct_core_dec(
		void splitAvailableBits(
		const int16_t total_bits, /* i : total available bits for TCX coding */
		const int16_t split_ratio, /* i : split ratio */
		const int16_t isSBAStereoMode, /* i : signal core coding for sba */
		const int16_t isSBAStereoMode, /* i : signal core coding for SBA */
		int16_t bits_ch0, / o : bits for channel 0 */
		int16_t bits_ch1 / o : bits for channel 1 */
		);
		@@ -2173,7 +2173,7 @@ void parse_stereo_from_bitstream(
		STEREO_MDCT_DEC_DATA_HANDLE hStereoMdct, /* i/o: MDCT stereo decoder structure */
		Decoder_State *sts, / i/o: decoder state structure */
		const int16_t mct_on, /* i : flag mct block (1) or stereo (0) */
		const int16_t isSBAStereoMode, /* i: flag core coding for sba */
		const int16_t isSBAStereoMode, /* i : flag core coding for SBA */
		Decoder_State st0, / i/o: decoder state structure for Bstr */
		int16_t ms_mask[NB_DIV][MAX_SFB] /* o : bandwise MS mask */
		);
		@@ -2508,8 +2508,7 @@ ivas_error stereo_memory_enc(
		const int32_t input_Fs, /* i : input sampling rate */
		const int16_t max_bwidth, /* i : maximum audio bandwidth */
		float tdm_last_ratio, / o : TD stereo last ratio */
		const IVAS_FORMAT ivas_format /* i : IVAS format */
		,
		const IVAS_FORMAT ivas_format, /* i : IVAS format */
		const int16_t nchan_transport /* i : number transport chans */

		);
		@@ -3363,8 +3362,8 @@ void ivas_dirac_dec_output_synthesis_process_subframe_psd_ls(
		float RealBuffer[][MAX_PARAM_SPATIAL_SUBFRAMES][CLDFB_NO_CHANNELS_MAX],/* i : LS signals */
		float ImagBuffer[][MAX_PARAM_SPATIAL_SUBFRAMES][CLDFB_NO_CHANNELS_MAX],/* i : LS signals */
		DIRAC_DEC_HANDLE hDirAC, /* i/o: DirAC handle */
		float *reference_power_smooth
		, float qualityBasedSmFactor
		float *reference_power_smooth,
		float qualityBasedSmFactor
		);

		void ivas_dirac_dec_get_response(
		@@ -3865,8 +3864,7 @@ void ivas_spar_dec_upmixer(
		/* MD module */
		ivas_error ivas_spar_md_enc_open(
		ivas_spar_md_enc_state_t *hMdEnc, / i/o: SPAR MD encoder handle */
		const ENCODER_CONFIG_HANDLE hEncoderConfig /* i : configuration structure */
		,
		const ENCODER_CONFIG_HANDLE hEncoderConfig, /* i : configuration structure */
		const int16_t sba_order /* i : Ambisonic (SBA) order */
		);

		@@ -3879,8 +3877,7 @@ ivas_error ivas_spar_md_enc_process(
		const ENCODER_CONFIG_HANDLE hEncoderConfig, /* i : configuration structure */
		ivas_spar_md_enc_in_buf_t *pIn_buf,
		BSTR_ENC_HANDLE hMetaData, /* i/o: MetaData handle */
		const int16_t dtx_silence_mode
		,
		const int16_t dtx_silence_mode,
		const int16_t sba_order /* i : Ambisonic (SBA) order */
		);

		@@ -4878,8 +4875,7 @@ void ivas_HRTF_binary_close(
		TDREND_HRFILT_FiltSet_t *hHrtfTD / i/o: TD renderer HRTF handle */
		);

		/! r: TD Renderer result code. /
		ivas_error DefaultBSplineModel(
		void DefaultBSplineModel(
		TDREND_HRFILT_FiltSet_t HrFiltSet_p, / o : Loaded HR filter set */
		const int32_t output_Fs /* i : Output sampling rate */
		);
		@@ -4892,7 +4888,7 @@ void ivas_td_binaural_close(
		BINAURAL_TD_OBJECT_RENDERER_HANDLE hBinRendererTd / i/o: TD binaural object renderer handle */
		);

		ivas_error ObjRenderIVASFrame(
		void ObjRenderIVASFrame(
		Decoder_Struct st_ivas, / i/o: IVAS decoder structure */
		float output[][L_FRAME48k], /* i/o: SCE channels / Binaural synthesis */
		const int16_t output_frame /* i : output frame length */
		@@ -4944,35 +4940,30 @@ ivas_error TDREND_REND_RenderSourceHRFilt(

		/* ----- Object renderer - sources ----- */

		/! r: TD Renderer result code. /
		ivas_error TDREND_MIX_SRC_SetPos(
		BINAURAL_TD_OBJECT_RENDERER_HANDLE hBinRendererTd, /* i/o: TD renderer handle */
		const int16_t SrcInd, /* i : Source index */
		const float Vec_p / i : Position vector */
		);

		/! r: TD Renderer result code. /
		ivas_error TDREND_MIX_SRC_SetDir(
		BINAURAL_TD_OBJECT_RENDERER_HANDLE hBinRendererTd, /* i/o: TD renderer handle */
		const int16_t SrcInd, /* i : Source index */
		const float Vec_p / i : Direction vector */
		);

		/! r: TD Renderer result code. /
		ivas_error TDREND_MIX_SRC_SetDirAtten(
		BINAURAL_TD_OBJECT_RENDERER_HANDLE hBinRendererTd, /* i/o: TD renderer handle */
		const int16_t SrcInd, /* i : Source index */
		const TDREND_DirAtten_t DirAtten_p / i : Directional attenuation specifier */
		);

		/! r: TD Renderer result code. /
		ivas_error TDREND_MIX_SRC_SetPlayState(
		BINAURAL_TD_OBJECT_RENDERER_HANDLE hBinRendererTd, /* i/o: TD renderer handle */
		const int16_t SrcInd, /* i : Source index */
		const TDREND_PlayStatus_t PlayStatus /* i : Play state */
		);


		void TDREND_SRC_REND_UpdateFiltersFromSpatialParams(
		BINAURAL_TD_OBJECT_RENDERER_HANDLE hBinRendererTd, /* i/o: TD renderer handle */
		TDREND_SRC_REND_t SrcRend_p, / i/o: Source object */
		@@ -4980,7 +4971,6 @@ void TDREND_SRC_REND_UpdateFiltersFromSpatialParams(
		const int32_t output_Fs /* i : Output sample rate */
		);

		/! r: TD Renderer result code. /
		ivas_error TDREND_SRC_Alloc(
		TDREND_SRC_t *Src_pp / i/o: Source */
		);
		@@ -5034,7 +5024,6 @@ int16_t TDREND_SPATIAL_EvalOrthonormOrient(

		/* ----- Object renderer - mix ----- */

		/! r: TD Renderer result code. /
		ivas_error TDREND_MIX_AddSrc(
		BINAURAL_TD_OBJECT_RENDERER_HANDLE hBinRendererTd, /* i/o: TD renderer handle */
		int16_t SrcInd, / o : Source index */
		@@ -5042,19 +5031,16 @@ ivas_error TDREND_MIX_AddSrc(
		const int32_t output_Fs /* i : Output sampling rate */
		);

		/! r: TD Renderer result code. /
		ivas_error TDREND_MIX_SetDistAttenModel(
		BINAURAL_TD_OBJECT_RENDERER_HANDLE hBinRendererTd, /* i/o: TD renderer handle */
		const TDREND_DistAttenModel_t DistAttenModel /* i : Distance attenuation model */
		);

		/! r: TD Renderer result code. /
		ivas_error TDREND_MIX_LIST_SetPos(
		void TDREND_MIX_LIST_SetPos(
		BINAURAL_TD_OBJECT_RENDERER_HANDLE hBinRendererTd, /* i/o: TD renderer handle */
		const float Pos_p / i : Listener's position */
		);

		/! r: TD Renderer result code. /
		ivas_error TDREND_MIX_LIST_SetOrient(
		BINAURAL_TD_OBJECT_RENDERER_HANDLE hBinRendererTd, /* i/o: TD renderer handle */
		const float FrontVec_p, / i : Listener's orientation front vector */
		@@ -5065,7 +5051,6 @@ void TDREND_MIX_Dealloc(
		BINAURAL_TD_OBJECT_RENDERER_HANDLE hBinRendererTd /* i/o: TD renderer handle */
		);

		/! r: TD Renderer result code. /
		ivas_error TDREND_MIX_Init(
		BINAURAL_TD_OBJECT_RENDERER_HANDLE hBinRendererTd, /* i/o: TD renderer handle */
		TDREND_HRFILT_FiltSet_t *hHrtfTD, / i/o: HRTF data (initialized in case of NULL) */
		@@ -5075,21 +5060,18 @@ ivas_error TDREND_MIX_Init(

		/* ----- Object renderer - sfx ----- */

		/! r: TD Renderer result code. /
		ivas_error TDREND_SFX_SpatBin_Initialize(
		SFX_SpatBin_t SfxSpatBin_p, / i/o: Spatial parameters handle */
		const int32_t output_Fs /* i : Output sampling rate */
		);

		/! r: TD Renderer result code. /
		ivas_error TDREND_SFX_SpatBin_SetParams(
		void TDREND_SFX_SpatBin_SetParams(
		SFX_SpatBin_t SfxSpatBin_p, / i/o: Spatial parameters struct to be updated */
		const SFX_SpatBin_Params_t NewParam_p, / i : New parameters struct */
		const int32_t output_Fs /* i : Output sample rate */
		);

		/! r: TD Renderer result code. /
		ivas_error TDREND_SFX_SpatBin_Execute_Main(
		void TDREND_SFX_SpatBin_Execute_Main(
		SFX_SpatBin_t SfxSpatBin_p, / i/o: Spatial parameters handle */
		const float InBuffer_p, / i : Input buffer */
		const int16_t output_frame, /* i : frame length */

lib_com/ivas_rom_com.c

+16 −12

Original line number	Diff line number	Diff line
		@@ -903,11 +903,11 @@ const ivas_spar_br_table_t ivas_spar_br_table_consts[IVAS_SPAR_BR_TABLE_LEN] =
		{
		/* When AGC is ON additional (AGC_BITS_PER_CH+1) bits may be taken from each core-coder channel
		so minimum core-coder bitrate per channel can be min core-coder bitrates as per the table - AGC_BITS_PER_CH */
		{ 24400, 0, 1, FB, 24000, 1, WYXZ, 1, 0,
		{ { 16400, 14850, 24350 } },
		{ 24400, 0, 1, FB, 24000, 1, WYXZ, 1, 0,{ { 16400, 14850, 24350 } },
		{ { 15, 1, 5, 1 },{ 15, 1, 3, 1 },{ 7, 1, 3, 1 } }, 0, 0, 0 },

		{ 32000, 0, 1, FB, 24000, 1, WYXZ, 1, 0,{{ 24000, 20450, 31950 }},{ { 21, 1, 5, 1 },{ 15, 1, 5, 1 },{ 15, 1, 3, 1 } }, 0, 0, 0 },
		{ 32000, 0, 1, FB, 24000, 1, WYXZ, 1, 0,{ { 24000, 20450, 31950 } },
		{ { 21, 1, 5, 1 },{ 15, 1, 5, 1 },{ 15, 1, 3, 1 } }, 0, 0, 0 },

		{ 48000, 0, 1, FB, 24000, 2, WYXZ, 0, 0,{ { 24000, 21000, 31950 },{ 16000, 15000, 20400 } },
		{ { 15, 7, 5, 1 },{ 15, 7, 3, 1 },{ 7, 7, 3, 1 } }, 1, 0, 0 },
		@@ -955,8 +955,6 @@ const ivas_spar_br_table_t ivas_spar_br_table_consts[IVAS_SPAR_BR_TABLE_LEN] =

		{ 512000, 0, 3, FB, 24000, 4, WYXZ, 0, 0,{ { 128000, 128000, 128000 },{ 128000, 128000, 128000 },{ 127200, 122550, 128000 },{ 76300, 73550, 128000 } }, // not yet optimized
		{ { 31, 11, 11, 1 },{ 1, 1, 1, 1 },{ 1, 1, 1, 1 } }, 1, 2, 0 },


		};

		const ivas_freq_models_t ivas_arith_pred_r_consts[TOTAL_PRED_QUANT_STRATS_ARITH] =
		@@ -3363,6 +3361,7 @@ const float ivas_mdft_coeff_cos_twid_960[IVAS_960_PT_LEN + 1] =
		0.00654493796735196f, 0.00490871880799808f, 0.00327248650652671f, 0.00163624544362412f,
		0.00000000000000000f
		};

		const float ivas_mdft_coeff_cos_twid_640[IVAS_640_PT_LEN + 1] =
		{
		1.00000000000000f, 0.999996988037278f, 0.999987952167257f, 0.999972892444367f,
		@@ -3524,9 +3523,10 @@ const float ivas_mdft_coeff_cos_twid_640[IVAS_640_PT_LEN +1] =
		0.0392598157590687f, 0.0368072229413588f, 0.0343544083996823f, 0.0319013869096110f,
		0.0294481732479634f, 0.0269947821927155f, 0.0245412285229123f, 0.0220875270185784f,
		0.0196336924606283f, 0.0171797396307788f, 0.0147256833114584f, 0.0122715382857199f,
		0.00981731933714973f, 0.00736304124977978f, 0.00490871880799808f, 0.00245436679646048f
		,0.00000000000000000f
		0.00981731933714973f, 0.00736304124977978f, 0.00490871880799808f, 0.00245436679646048f,
		0.00000000000000000f
		};

		const float ivas_mdft_coeff_cos_twid_320[IVAS_320_PT_LEN + 1] =
		{
		1.00000000000000f, 0.999987952167257f, 0.999951808959328f, 0.999891571247108f,
		@@ -3608,9 +3608,10 @@ const float ivas_mdft_coeff_cos_twid_320[IVAS_320_PT_LEN + 1] =
		0.0784590957278450f, 0.0735645635996675f, 0.0686682588843738f, 0.0637702995616845f,
		0.0588708036511890f, 0.0539698892095020f, 0.0490676743274181f, 0.0441642771270675f,
		0.0392598157590687f, 0.0343544083996823f, 0.0294481732479634f, 0.0245412285229123f,
		0.0196336924606283f, 0.0147256833114584f, 0.00981731933714973f, 0.00490871880799808f
		,0.0000000000000000f
		0.0196336924606283f, 0.0147256833114584f, 0.00981731933714973f, 0.00490871880799808f,
		0.0000000000000000f
		};

		const float ivas_mdft_coeff_cos_twid_240[IVAS_240_PT_LEN + 1] =
		{
		1.0000000000f, 0.9999785817f, 0.9999143276f, 0.9998072405f, 0.9996573250f, 0.9994645875f,
		@@ -3685,6 +3686,7 @@ const float ivas_mdft_coeff_cos_twid_160[IVAS_160_PT_LEN + 1] =
		0.0980171403f, 0.0882423705f, 0.0784590957f, 0.0686682589f, 0.0588708037f, 0.0490676743f,
		0.0392598158f, 0.0294481732f, 0.0196336925f, 0.0098173193f, 0.000000000f
		};

		const float ivas_mdft_coeff_cos_twid_80[IVAS_80_PT_LEN + 1] =
		{
		1.0000000000f, 0.9998072405f, 0.9992290362f, 0.9982656102f, 0.9969173337f, 0.9951847267f,
		@@ -3702,6 +3704,7 @@ const float ivas_mdft_coeff_cos_twid_80[IVAS_80_PT_LEN + 1] =
		0.1564344650f, 0.1370123417f, 0.1175373975f, 0.0980171403f, 0.0784590957f, 0.0588708037f,
		0.0392598158f, 0.0196336925f, 0.000000000f
		};

		const float ivas_mdft_coeff_cos_twid_40[IVAS_40_PT_LEN + 1] =
		{
		1.0000000000f, 0.9992290362f, 0.9969173337f, 0.9930684570f, 0.9876883406f, 0.9807852804f,
		@@ -3712,6 +3715,7 @@ const float ivas_mdft_coeff_cos_twid_40[IVAS_40_PT_LEN + 1] =
		0.3826834324f, 0.3461170571f, 0.3090169944f, 0.2714404499f, 0.2334453639f, 0.1950903220f,
		0.1564344650f, 0.1175373975f, 0.0784590957f, 0.0392598158f, 0.000000000f
		};

		const float ivas_sin_twiddle_480[IVAS_480_PT_LEN >> 1] =
		{
		-0.000818122995607253f, -0.00736304124977957f, -0.0139076440957708f, -0.0204516511845773f,