Commit 244f67a0 authored by vaclav's avatar vaclav
Browse files

Merge remote-tracking branch 'remotes/origin/main' into cleanup_20250422

parents 2e72e8e0 f0adfdc9
Loading
Loading
Loading
Loading
Loading
+25 −33
Original line number Diff line number Diff line
@@ -69,22 +69,10 @@ static
#endif
    int32_t frame = 0; /* Counter of frames */

#ifdef DEBUGGING
#define MIN_NUM_BITS_ACTIVE_FRAME 56
#define NUM_BITS_SID_IVAS_5K2     104
#define MAX_OUTPUT_PCM_BUFFER_SIZE ( ( IVAS_MAX_OUTPUT_CHANNELS + IVAS_MAX_NUM_OBJECTS ) * IVAS_MAX_FRAME_SIZE )

#define ORIENT_TRK_NONE        ( 0 )
#define ORIENT_TRK_REF         ( 1 )
#define ORIENT_TRK_AVG         ( 2 )
#define ORIENT_TRK_REF_VEC     ( 3 )
#define ORIENT_TRK_REF_VEC_LEV ( 4 )

#ifdef VARIABLE_SPEED_DECODING
#define VARIABLE_SPEED_FETCH_FRAMESIZE_MS 20
#endif
#define JBM_FRONTEND_FETCH_FRAMESIZE_MS 20
#define HEADROTATION_FETCH_FRAMESIZE_MS 5
#define DEFAULT_FETCH_FRAMESIZE_MS      20


/*------------------------------------------------------------------------------------------*
@@ -129,7 +117,7 @@ typedef struct
    IVAS_DEC_INPUT_FORMAT inputFormat;
    bool customLsOutputEnabled;
    char *customLsSetupFilename;
    int16_t orientation_tracking;
    IVAS_HEAD_ORIENT_TRK_T orientation_tracking;
    bool non_diegetic_pan_enabled;
    float non_diegetic_pan_gain;
    bool renderConfigEnabled;
@@ -183,7 +171,7 @@ typedef struct
static bool parseCmdlIVAS_dec( int16_t argc, char **argv, DecArguments *arg );
static void usage_dec( void );
static ivas_error decodeG192( DecArguments arg, BS_READER_HANDLE hBsReader, IVAS_DEC_HRTF_BINARY_WRAPPER *hHrtf, RotFileReader *headRotReader, RotFileReader *externalOrientationFileReader, RotFileReader *refRotReader, Vector3PairFileReader *referenceVectorReader, ObjectEditFileReader *objectEditFileReader, ISAR_SPLIT_REND_BITS_DATA *splitRendBits, IVAS_DEC_HANDLE hIvasDec, int16_t *pcmBuf );
static ivas_error decodeVoIP( DecArguments arg, BS_READER_HANDLE hBsReader, IVAS_DEC_HRTF_BINARY_WRAPPER *hHrtf, RotFileReader *headRotReader, RotFileReader *externalOrientationFileReader, RotFileReader *refRotReader, Vector3PairFileReader *referenceVectorReader, ObjectEditFileReader *objectEditFileReader, IVAS_DEC_HANDLE hIvasDec );
static ivas_error decodeVoIP( DecArguments arg, BS_READER_HANDLE hBsReader, IVAS_DEC_HRTF_BINARY_WRAPPER *hHrtf, RotFileReader *headRotReader, RotFileReader *externalOrientationFileReader, RotFileReader *refRotReader, Vector3PairFileReader *referenceVectorReader, ObjectEditFileReader *objectEditFileReader, IVAS_DEC_HANDLE hIvasDec, int16_t *pcmBuf );
static ivas_error load_hrtf_from_file( IVAS_DEC_HRTF_BINARY_WRAPPER *hHrtfBinary, IVAS_DEC_HANDLE hIvasDec, const IVAS_AUDIO_CONFIG OutputConfig, const int32_t output_Fs );
#ifdef DEBUGGING
static ivas_error printBitstreamInfoVoip( DecArguments arg, BS_READER_HANDLE hBsReader, IVAS_DEC_HANDLE hIvasDec );
@@ -359,7 +347,7 @@ int main(
        }

        /* sanity check */
        if ( arg.orientation_tracking != ORIENT_TRK_REF )
        if ( arg.orientation_tracking != IVAS_HEAD_ORIENT_TRK_REF )
        {
            fprintf( stderr, "\nError: Reference rotation file can be used in '-otr ref' mode only.\n\n" );
            goto cleanup;
@@ -386,7 +374,7 @@ int main(
        }

        /* sanity check */
        if ( arg.orientation_tracking != ORIENT_TRK_REF_VEC && arg.orientation_tracking != ORIENT_TRK_REF_VEC_LEV )
        if ( arg.orientation_tracking != IVAS_HEAD_ORIENT_TRK_REF_VEC && arg.orientation_tracking != IVAS_HEAD_ORIENT_TRK_REF_VEC_LEV )
        {
            fprintf( stderr, "\nError: Reference trajectory file can be used in '-otr ref_vec' or '-otr ref_vec_lev' mode only.\n\n" );
            goto cleanup;
@@ -764,7 +752,7 @@ int main(

    if ( arg.voipMode )
    {
        error = decodeVoIP( arg, hBsReader, &hHrtfBinary, headRotReader, externalOrientationFileReader, refRotReader, referenceVectorReader, objectEditFileReader, hIvasDec );
        error = decodeVoIP( arg, hBsReader, &hHrtfBinary, headRotReader, externalOrientationFileReader, refRotReader, referenceVectorReader, objectEditFileReader, hIvasDec, pcmBuf );
    }
    else
    {
@@ -962,6 +950,10 @@ static bool parseCmdlIVAS_dec(
    int16_t i;
    char argv_to_upper[FILENAME_MAX];

    /*-----------------------------------------------------------------*
     * Set default values
     *-----------------------------------------------------------------*/

#ifdef DEBUGGING
    float ftmp;

@@ -970,7 +962,7 @@ static bool parseCmdlIVAS_dec(
    arg->agcBitstream = NULL;
#endif
#endif
    arg->output_Fs = 48000;
    arg->output_Fs = IVAS_MAX_SAMPLING_RATE;
    arg->outputConfig = IVAS_AUDIO_CONFIG_MONO;
    arg->decMode = IVAS_DEC_MODE_IVAS;
    arg->quietModeEnabled = false;
@@ -980,7 +972,7 @@ static bool parseCmdlIVAS_dec(

    arg->enableHeadRotation = false;
    arg->headrotTrajFileName = NULL;
    arg->orientation_tracking = ORIENT_TRK_NONE;
    arg->orientation_tracking = IVAS_HEAD_ORIENT_TRK_NONE;
    arg->enableReferenceRotation = false;
    arg->headrotTrajFileName = NULL;
    arg->enableReferenceVectorTracking = false;
@@ -1180,9 +1172,9 @@ static bool parseCmdlIVAS_dec(
                }
                arg->tsmScale = (uint16_t) tmp;

                if ( arg->tsmScale < 50 || arg->tsmScale > 150 )
                if ( arg->tsmScale < IVAS_TIME_SCALE_MIN || arg->tsmScale > IVAS_TIME_SCALE_MAX )
                {
                    fprintf( stderr, "Error: Scaling factor value must be 50 <= fac <= 150!\n\n" );
                    fprintf( stderr, "Error: Scaling factor value must be IVAS_TIME_SCALE_MIN <= fac <= IVAS_TIME_SCALE_MAX!\n\n" );
                    usage_dec();
                    return false;
                }
@@ -1255,23 +1247,23 @@ static bool parseCmdlIVAS_dec(

            if ( strcmp( argv_to_upper, "NONE" ) == 0 )
            {
                arg->orientation_tracking = ORIENT_TRK_NONE;
                arg->orientation_tracking = IVAS_HEAD_ORIENT_TRK_NONE;
            }
            else if ( strcmp( argv_to_upper, "REF" ) == 0 )
            {
                arg->orientation_tracking = ORIENT_TRK_REF;
                arg->orientation_tracking = IVAS_HEAD_ORIENT_TRK_REF;
            }
            else if ( strcmp( argv_to_upper, "AVG" ) == 0 )
            {
                arg->orientation_tracking = ORIENT_TRK_AVG;
                arg->orientation_tracking = IVAS_HEAD_ORIENT_TRK_AVG;
            }
            else if ( strcmp( argv_to_upper, "REF_VEC" ) == 0 )
            {
                arg->orientation_tracking = ORIENT_TRK_REF_VEC;
                arg->orientation_tracking = IVAS_HEAD_ORIENT_TRK_REF_VEC;
            }
            else if ( strcmp( argv_to_upper, "REF_VEC_LEV" ) == 0 )
            {
                arg->orientation_tracking = ORIENT_TRK_REF_VEC_LEV;
                arg->orientation_tracking = IVAS_HEAD_ORIENT_TRK_REF_VEC_LEV;
            }
            else
            {
@@ -1666,7 +1658,7 @@ static void usage_dec( void )
#ifdef DEBUGGING
#ifdef VARIABLE_SPEED_DECODING
    fprintf( stdout, "-VS fac             : Variable Speed mode: change speed of playout fac as integer in percent.\n" );
    fprintf( stdout, "                      50 <= fac <= 150; fac<100 faster, fac>100 slower\n" );
    fprintf( stdout, "                      IVAS_TIME_SCALE_MIN <= fac <= IVAS_TIME_SCALE_MAX; fac<100 faster, fac>100 slower\n" );
#endif
#endif
    fprintf( stdout, "-fr L               : render frame size in ms L=(5,10,20), default is 20)\n" );
@@ -2988,14 +2980,15 @@ static ivas_error decodeVoIP(
    RotFileReader *refRotReader,
    Vector3PairFileReader *referenceVectorReader,
    ObjectEditFileReader *objectEditFileReader,
    IVAS_DEC_HANDLE hIvasDec )
    IVAS_DEC_HANDLE hIvasDec,
    int16_t *pcmBuf )
{
    bool decodingFailed = true; /* Assume failure until cleanup is reached without errors */
    ivas_error error = IVAS_ERR_OK;

    uint32_t nextPacketRcvTime_ms = 0;
    uint32_t systemTime_ms = 0;
    uint32_t systemTimeInc_ms = (uint32_t) JBM_FRONTEND_FETCH_FRAMESIZE_MS;
    uint32_t systemTimeInc_ms = (uint32_t) ( 1000 / IVAS_NUM_FRAMES_PER_SEC );
    int32_t nFramesFed = 0;

    uint8_t au[( IVAS_MAX_BITS_PER_FRAME + 7 ) >> 3];
@@ -3009,7 +3002,6 @@ static ivas_error decodeVoIP(
    MasaFileWriter *masaWriter = NULL;
    uint16_t numObj = 0;

    int16_t pcmBuf[MAX_OUTPUT_PCM_BUFFER_SIZE];
    AudioFileWriter *afWriter = NULL;
#ifdef SUPPORT_JBM_TRACEFILE
    JbmTraceFileWriter *jbmTraceWriter = NULL;
+22 −26
Original line number Diff line number Diff line
@@ -154,7 +154,6 @@ typedef struct
 * Local functions prototypes
 *------------------------------------------------------------------------------------------*/

static void initArgStruct( EncArguments *arg );
static bool parseCmdlIVAS_enc( int16_t argc, char *argv[], EncArguments *arg );
static void usage_enc( void );
static bool readBandwidth( FILE *file, IVAS_ENC_BANDWIDTH *bandwidth, int32_t *bandwidthFrameCounter );
@@ -207,8 +206,6 @@ int main(
    reset_mem( USE_BYTES );
#endif

    initArgStruct( &arg );

    /*------------------------------------------------------------------------------------------*
     * Parse command-line arguments
     *------------------------------------------------------------------------------------------*/
@@ -878,14 +875,29 @@ cleanup:
 * Local functions
 *-------------------------------------------------------------------*/

#define IVAS_DEFAULT_AGC ( 0 )
/*---------------------------------------------------------------------*
 * parseCmdlIVAS_enc()
 *
 * Encoder command-line parsing
 *---------------------------------------------------------------------*/

static void initArgStruct( EncArguments *arg )
static bool parseCmdlIVAS_enc(
    int16_t argc,
    char *argv[],
    EncArguments *arg )
{
    /* Set default values here */
    int16_t i, j;
    char argv_to_upper[FILENAME_MAX];
    char stmp[FILENAME_MAX];
    int32_t tmp;

    /*-----------------------------------------------------------------*
     * Set default values
     *-----------------------------------------------------------------*/

    arg->inputWavFilename = NULL;
    arg->outputBitstreamFilename = NULL;
    arg->inputFs = 0;
    arg->inputFs = IVAS_MAX_SAMPLING_RATE;
    arg->inputFormat = IVAS_ENC_INPUT_MONO;
    arg->is_binaural = false;
    arg->inputFormatConfig.stereoToMonoDownmix = false;
@@ -919,25 +931,9 @@ static void initArgStruct( EncArguments *arg )
#endif
    arg->pca = false;

    return;
}


/*---------------------------------------------------------------------*
 * parseCmdlIVAS_enc()
 *
 * Encoder command-line parsing
 *---------------------------------------------------------------------*/

static bool parseCmdlIVAS_enc(
    int16_t argc,
    char *argv[],
    EncArguments *arg )
{
    int16_t i, j;
    char argv_to_upper[FILENAME_MAX];
    char stmp[FILENAME_MAX];
    int32_t tmp;
    /*-----------------------------------------------------------------*
     * Initialization
     *-----------------------------------------------------------------*/

    if ( argc < 5 )
    {
+1 −1
Original line number Diff line number Diff line
@@ -945,7 +945,7 @@ int main(

    memset( outBuffer.data, 0, outBuffer.config.numSamplesPerChannel * outBuffer.config.numChannels * sizeof( float ) );

    bitsBufferSize = SPLIT_REND_BITS_BUFF_SIZE;
    bitsBufferSize = ISAR_SPLIT_REND_BITS_BUFF_SIZE;

    if ( bitsBufferSize > 0 )
    {
+3 −3
Original line number Diff line number Diff line
@@ -168,7 +168,7 @@ typedef struct
    char externalOrientationFilePath[RENDERER_MAX_CLI_ARG_LENGTH];
    char customHrtfFilePath[RENDERER_MAX_CLI_ARG_LENGTH];
    char renderConfigFilePath[RENDERER_MAX_CLI_ARG_LENGTH];
    int8_t orientation_tracking;
    IVAS_HEAD_ORIENT_TRK_T orientation_tracking;
    int16_t Opt_Headrotation;
    int16_t Opt_ExternalOrientation;
    int16_t nonDiegeticPan;
@@ -1436,7 +1436,7 @@ int main(

    if ( is_split_pre_rend_mode( &args ) )
    {
        bitsBufferSize = SPLIT_REND_BITS_BUFF_SIZE;
        bitsBufferSize = ISAR_SPLIT_REND_BITS_BUFF_SIZE;
    }
    else
    {
@@ -2268,7 +2268,7 @@ static bool parseRenderFramesize(

static bool parseOrientationTracking(
    char *value,
    int8_t *orientation_tracking )
    IVAS_HEAD_ORIENT_TRK_T *orientation_tracking )
{
    to_upper( value );

+19 −15
Original line number Diff line number Diff line
@@ -44,24 +44,27 @@
 * Common API constants
 *----------------------------------------------------------------------------------*/

#define IVAS_NUM_FRAMES_PER_SEC 50
#define IVAS_MAX_FRAME_SIZE     ( 48000 / IVAS_NUM_FRAMES_PER_SEC )
#define IVAS_MAX_BITS_PER_FRAME ( 512000 / IVAS_NUM_FRAMES_PER_SEC )
#define IVAS_NUM_FRAMES_PER_SEC 50                                                   /* number of frames per second */
#define IVAS_MAX_SAMPLING_RATE  48000                                                /* maximum supported sampling rate in Hz */
#define IVAS_MAX_FRAME_SIZE     ( IVAS_MAX_SAMPLING_RATE / IVAS_NUM_FRAMES_PER_SEC ) /* maximum frame buffer length (per channel) in samples */
#define IVAS_MAX_BITS_PER_FRAME ( 512000 / IVAS_NUM_FRAMES_PER_SEC )                 /* maximum bits per frame; corresponds to maximum bitrate of 512 kbps */

#define IVAS_MAX_NUM_OBJECTS       4
#define IVAS_MAX_INPUT_CHANNELS    16
#define IVAS_MAX_OUTPUT_CHANNELS   16
#define IVAS_MAX_OUTPUT_CHANNELS   16 /* Note: there is an exception for OSBA and EXT otuput where it can be 20 (HOA3 + 4 ISM channels) */
#define IVAS_CLDFB_NO_COL_MAX      16
#define IVAS_CLDFB_NO_CHANNELS_MAX 60

/* Renderer constants */
#define IVAS_MAX_PARAM_SPATIAL_SUBFRAMES 4
#define IVAS_ROOM_ABS_COEFF              6
#define IVAS_REVERB_PREDELAY_MAX         20 /* Max input delay for reverb module */
#define IVAS_ER_LIST_HEIGHT              1.6f

/* Maximum buffer length (per channel) in samples */
#define MAX_BUFFER_LENGTH_PER_CHANNEL ( L_FRAME48k )
/* JBM constants for adaptive-playout */
#define IVAS_TIME_SCALE_MIN 50  /* min. time-scaling [%] */
#define IVAS_TIME_SCALE_MAX 150 /* max. time-scaling [%] */

/* Frame size required when rendering to binaural */
#define BINAURAL_RENDERING_FRAME_SIZE_MS 5

/*----------------------------------------------------------------------------------*
 * Common API enum for output audio configurations
@@ -174,12 +177,12 @@ typedef enum
typedef struct ivas_masa_metadata_frame_struct *IVAS_MASA_METADATA_HANDLE;
typedef struct ivas_masa_decoder_ext_out_meta_struct *IVAS_MASA_DECODER_EXT_OUT_META_HANDLE;

typedef struct TDREND_HRFILT_FiltSet_struct *IVAS_DEC_HRTF_HANDLE;
typedef struct ivas_hrtf_TDREND_HRFILT_FiltSet_struct *IVAS_DEC_HRTF_HANDLE;
typedef struct ivas_hrtf_crend_structure *IVAS_DEC_HRTF_CREND_HANDLE;
typedef struct ivas_hrtfs_fastconv_struct *IVAS_DEC_HRTF_FASTCONV_HANDLE;
typedef struct ivas_hrtfs_parambin_struct *IVAS_DEC_HRTF_PARAMBIN_HANDLE;
typedef struct ivas_hrtfs_statistics_struct *IVAS_DEC_HRTF_STATISTICS_HANDLE;
typedef struct cldfb_filter_bank_struct *IVAS_CLDFB_FILTER_BANK_HANDLE;
typedef struct ivas_hrtf_fastconv_struct *IVAS_DEC_HRTF_FASTCONV_HANDLE;
typedef struct ivas_hrtf_parambin_struct *IVAS_DEC_HRTF_PARAMBIN_HANDLE;
typedef struct ivas_hrtf_statistics_struct *IVAS_DEC_HRTF_STATISTICS_HANDLE;
typedef struct ivas_cldfb_filter_bank_struct *IVAS_CLDFB_FILTER_BANK_HANDLE;

typedef struct ivas_LS_setup_custom *IVAS_LSSETUP_CUSTOM_HANDLE;
typedef struct ivas_LS_setup_custom IVAS_LSSETUP_CUSTOM_STRUCT;
@@ -228,7 +231,7 @@ typedef enum _ivas_binaural_renderer_type
#define ISAR_MAX_SPLIT_REND_BITRATE                   768000
#define ISAR_MAX_SPLIT_REND_BITS_BUFFER_SIZE_IN_BYTES ( ( ( (int32_t) ISAR_MAX_SPLIT_REND_BITRATE / IVAS_NUM_FRAMES_PER_SEC ) + 7 ) >> 3 )
#define ISAR_SPLIT_REND_ADDITIONAL_BYTES_TO_READ      1
#define SPLIT_REND_BITS_BUFF_SIZE                     ( ISAR_MAX_SPLIT_REND_BITS_BUFFER_SIZE_IN_BYTES + ISAR_SPLIT_REND_ADDITIONAL_BYTES_TO_READ )
#define ISAR_SPLIT_REND_BITS_BUFF_SIZE                ( ISAR_MAX_SPLIT_REND_BITS_BUFFER_SIZE_IN_BYTES + ISAR_SPLIT_REND_ADDITIONAL_BYTES_TO_READ )

typedef enum
{
@@ -306,8 +309,8 @@ typedef enum
    IVAS_RENDER_TYPE_OVERRIDE_FASTCONV

} IVAS_RENDER_TYPE_OVERRIDE;
#endif

#endif
typedef struct _IVAS_ROOM_ACOUSTICS_CONFIG
{
    int16_t nBands;                                   /* Number of frequency bands for which reverb properties are provided, integer, range [2..256]        */
@@ -350,6 +353,7 @@ typedef struct
{
    IVAS_REND_AudioBufferConfig config;
    float *data;

} IVAS_REND_AudioBuffer;

#endif /* COMMON_API_TYPES_H */
Loading