LCOV - code coverage report
Current view: top level - libavcodec - wmavoice.c (source / functions) Hit Total Coverage
Test: coverage.info Lines: 660 747 88.4 %
Date: 2018-05-20 11:54:08 Functions: 28 29 96.6 %

          Line data    Source code
       1             : /*
       2             :  * Windows Media Audio Voice decoder.
       3             :  * Copyright (c) 2009 Ronald S. Bultje
       4             :  *
       5             :  * This file is part of FFmpeg.
       6             :  *
       7             :  * FFmpeg is free software; you can redistribute it and/or
       8             :  * modify it under the terms of the GNU Lesser General Public
       9             :  * License as published by the Free Software Foundation; either
      10             :  * version 2.1 of the License, or (at your option) any later version.
      11             :  *
      12             :  * FFmpeg is distributed in the hope that it will be useful,
      13             :  * but WITHOUT ANY WARRANTY; without even the implied warranty of
      14             :  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
      15             :  * Lesser General Public License for more details.
      16             :  *
      17             :  * You should have received a copy of the GNU Lesser General Public
      18             :  * License along with FFmpeg; if not, write to the Free Software
      19             :  * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
      20             :  */
      21             : 
      22             : /**
      23             :  * @file
      24             :  * @brief Windows Media Audio Voice compatible decoder
      25             :  * @author Ronald S. Bultje <rsbultje@gmail.com>
      26             :  */
      27             : 
      28             : #include <math.h>
      29             : 
      30             : #include "libavutil/channel_layout.h"
      31             : #include "libavutil/float_dsp.h"
      32             : #include "libavutil/mem.h"
      33             : #include "libavutil/thread.h"
      34             : #include "avcodec.h"
      35             : #include "internal.h"
      36             : #include "get_bits.h"
      37             : #include "put_bits.h"
      38             : #include "wmavoice_data.h"
      39             : #include "celp_filters.h"
      40             : #include "acelp_vectors.h"
      41             : #include "acelp_filters.h"
      42             : #include "lsp.h"
      43             : #include "dct.h"
      44             : #include "rdft.h"
      45             : #include "sinewin.h"
      46             : 
      47             : #define MAX_BLOCKS           8   ///< maximum number of blocks per frame
      48             : #define MAX_LSPS             16  ///< maximum filter order
      49             : #define MAX_LSPS_ALIGN16     16  ///< same as #MAX_LSPS; needs to be multiple
      50             :                                  ///< of 16 for ASM input buffer alignment
      51             : #define MAX_FRAMES           3   ///< maximum number of frames per superframe
      52             : #define MAX_FRAMESIZE        160 ///< maximum number of samples per frame
      53             : #define MAX_SIGNAL_HISTORY   416 ///< maximum excitation signal history
      54             : #define MAX_SFRAMESIZE       (MAX_FRAMESIZE * MAX_FRAMES)
      55             :                                  ///< maximum number of samples per superframe
      56             : #define SFRAME_CACHE_MAXSIZE 256 ///< maximum cache size for frame data that
      57             :                                  ///< was split over two packets
      58             : #define VLC_NBITS            6   ///< number of bits to read per VLC iteration
      59             : 
      60             : /**
      61             :  * Frame type VLC coding.
      62             :  */
      63             : static VLC frame_type_vlc;
      64             : 
      65             : /**
      66             :  * Adaptive codebook types.
      67             :  */
      68             : enum {
      69             :     ACB_TYPE_NONE       = 0, ///< no adaptive codebook (only hardcoded fixed)
      70             :     ACB_TYPE_ASYMMETRIC = 1, ///< adaptive codebook with per-frame pitch, which
      71             :                              ///< we interpolate to get a per-sample pitch.
      72             :                              ///< Signal is generated using an asymmetric sinc
      73             :                              ///< window function
      74             :                              ///< @note see #wmavoice_ipol1_coeffs
      75             :     ACB_TYPE_HAMMING    = 2  ///< Per-block pitch with signal generation using
      76             :                              ///< a Hamming sinc window function
      77             :                              ///< @note see #wmavoice_ipol2_coeffs
      78             : };
      79             : 
      80             : /**
      81             :  * Fixed codebook types.
      82             :  */
      83             : enum {
      84             :     FCB_TYPE_SILENCE    = 0, ///< comfort noise during silence
      85             :                              ///< generated from a hardcoded (fixed) codebook
      86             :                              ///< with per-frame (low) gain values
      87             :     FCB_TYPE_HARDCODED  = 1, ///< hardcoded (fixed) codebook with per-block
      88             :                              ///< gain values
      89             :     FCB_TYPE_AW_PULSES  = 2, ///< Pitch-adaptive window (AW) pulse signals,
      90             :                              ///< used in particular for low-bitrate streams
      91             :     FCB_TYPE_EXC_PULSES = 3, ///< Innovation (fixed) codebook pulse sets in
      92             :                              ///< combinations of either single pulses or
      93             :                              ///< pulse pairs
      94             : };
      95             : 
      96             : /**
      97             :  * Description of frame types.
      98             :  */
      99             : static const struct frame_type_desc {
     100             :     uint8_t n_blocks;     ///< amount of blocks per frame (each block
     101             :                           ///< (contains 160/#n_blocks samples)
     102             :     uint8_t log_n_blocks; ///< log2(#n_blocks)
     103             :     uint8_t acb_type;     ///< Adaptive codebook type (ACB_TYPE_*)
     104             :     uint8_t fcb_type;     ///< Fixed codebook type (FCB_TYPE_*)
     105             :     uint8_t dbl_pulses;   ///< how many pulse vectors have pulse pairs
     106             :                           ///< (rather than just one single pulse)
     107             :                           ///< only if #fcb_type == #FCB_TYPE_EXC_PULSES
     108             : } frame_descs[17] = {
     109             :     { 1, 0, ACB_TYPE_NONE,       FCB_TYPE_SILENCE,    0 },
     110             :     { 2, 1, ACB_TYPE_NONE,       FCB_TYPE_HARDCODED,  0 },
     111             :     { 2, 1, ACB_TYPE_ASYMMETRIC, FCB_TYPE_AW_PULSES,  0 },
     112             :     { 2, 1, ACB_TYPE_ASYMMETRIC, FCB_TYPE_EXC_PULSES, 2 },
     113             :     { 2, 1, ACB_TYPE_ASYMMETRIC, FCB_TYPE_EXC_PULSES, 5 },
     114             :     { 4, 2, ACB_TYPE_ASYMMETRIC, FCB_TYPE_EXC_PULSES, 0 },
     115             :     { 4, 2, ACB_TYPE_ASYMMETRIC, FCB_TYPE_EXC_PULSES, 2 },
     116             :     { 4, 2, ACB_TYPE_ASYMMETRIC, FCB_TYPE_EXC_PULSES, 5 },
     117             :     { 2, 1, ACB_TYPE_HAMMING,    FCB_TYPE_EXC_PULSES, 0 },
     118             :     { 2, 1, ACB_TYPE_HAMMING,    FCB_TYPE_EXC_PULSES, 2 },
     119             :     { 2, 1, ACB_TYPE_HAMMING,    FCB_TYPE_EXC_PULSES, 5 },
     120             :     { 4, 2, ACB_TYPE_HAMMING,    FCB_TYPE_EXC_PULSES, 0 },
     121             :     { 4, 2, ACB_TYPE_HAMMING,    FCB_TYPE_EXC_PULSES, 2 },
     122             :     { 4, 2, ACB_TYPE_HAMMING,    FCB_TYPE_EXC_PULSES, 5 },
     123             :     { 8, 3, ACB_TYPE_HAMMING,    FCB_TYPE_EXC_PULSES, 0 },
     124             :     { 8, 3, ACB_TYPE_HAMMING,    FCB_TYPE_EXC_PULSES, 2 },
     125             :     { 8, 3, ACB_TYPE_HAMMING,    FCB_TYPE_EXC_PULSES, 5 }
     126             : };
     127             : 
     128             : /**
     129             :  * WMA Voice decoding context.
     130             :  */
     131             : typedef struct WMAVoiceContext {
     132             :     /**
     133             :      * @name Global values specified in the stream header / extradata or used all over.
     134             :      * @{
     135             :      */
     136             :     GetBitContext gb;             ///< packet bitreader. During decoder init,
     137             :                                   ///< it contains the extradata from the
     138             :                                   ///< demuxer. During decoding, it contains
     139             :                                   ///< packet data.
     140             :     int8_t vbm_tree[25];          ///< converts VLC codes to frame type
     141             : 
     142             :     int spillover_bitsize;        ///< number of bits used to specify
     143             :                                   ///< #spillover_nbits in the packet header
     144             :                                   ///< = ceil(log2(ctx->block_align << 3))
     145             :     int history_nsamples;         ///< number of samples in history for signal
     146             :                                   ///< prediction (through ACB)
     147             : 
     148             :     /* postfilter specific values */
     149             :     int do_apf;                   ///< whether to apply the averaged
     150             :                                   ///< projection filter (APF)
     151             :     int denoise_strength;         ///< strength of denoising in Wiener filter
     152             :                                   ///< [0-11]
     153             :     int denoise_tilt_corr;        ///< Whether to apply tilt correction to the
     154             :                                   ///< Wiener filter coefficients (postfilter)
     155             :     int dc_level;                 ///< Predicted amount of DC noise, based
     156             :                                   ///< on which a DC removal filter is used
     157             : 
     158             :     int lsps;                     ///< number of LSPs per frame [10 or 16]
     159             :     int lsp_q_mode;               ///< defines quantizer defaults [0, 1]
     160             :     int lsp_def_mode;             ///< defines different sets of LSP defaults
     161             :                                   ///< [0, 1]
     162             : 
     163             :     int min_pitch_val;            ///< base value for pitch parsing code
     164             :     int max_pitch_val;            ///< max value + 1 for pitch parsing
     165             :     int pitch_nbits;              ///< number of bits used to specify the
     166             :                                   ///< pitch value in the frame header
     167             :     int block_pitch_nbits;        ///< number of bits used to specify the
     168             :                                   ///< first block's pitch value
     169             :     int block_pitch_range;        ///< range of the block pitch
     170             :     int block_delta_pitch_nbits;  ///< number of bits used to specify the
     171             :                                   ///< delta pitch between this and the last
     172             :                                   ///< block's pitch value, used in all but
     173             :                                   ///< first block
     174             :     int block_delta_pitch_hrange; ///< 1/2 range of the delta (full range is
     175             :                                   ///< from -this to +this-1)
     176             :     uint16_t block_conv_table[4]; ///< boundaries for block pitch unit/scale
     177             :                                   ///< conversion
     178             : 
     179             :     /**
     180             :      * @}
     181             :      *
     182             :      * @name Packet values specified in the packet header or related to a packet.
     183             :      *
     184             :      * A packet is considered to be a single unit of data provided to this
     185             :      * decoder by the demuxer.
     186             :      * @{
     187             :      */
     188             :     int spillover_nbits;          ///< number of bits of the previous packet's
     189             :                                   ///< last superframe preceding this
     190             :                                   ///< packet's first full superframe (useful
     191             :                                   ///< for re-synchronization also)
     192             :     int has_residual_lsps;        ///< if set, superframes contain one set of
     193             :                                   ///< LSPs that cover all frames, encoded as
     194             :                                   ///< independent and residual LSPs; if not
     195             :                                   ///< set, each frame contains its own, fully
     196             :                                   ///< independent, LSPs
     197             :     int skip_bits_next;           ///< number of bits to skip at the next call
     198             :                                   ///< to #wmavoice_decode_packet() (since
     199             :                                   ///< they're part of the previous superframe)
     200             : 
     201             :     uint8_t sframe_cache[SFRAME_CACHE_MAXSIZE + AV_INPUT_BUFFER_PADDING_SIZE];
     202             :                                   ///< cache for superframe data split over
     203             :                                   ///< multiple packets
     204             :     int sframe_cache_size;        ///< set to >0 if we have data from an
     205             :                                   ///< (incomplete) superframe from a previous
     206             :                                   ///< packet that spilled over in the current
     207             :                                   ///< packet; specifies the amount of bits in
     208             :                                   ///< #sframe_cache
     209             :     PutBitContext pb;             ///< bitstream writer for #sframe_cache
     210             : 
     211             :     /**
     212             :      * @}
     213             :      *
     214             :      * @name Frame and superframe values
     215             :      * Superframe and frame data - these can change from frame to frame,
     216             :      * although some of them do in that case serve as a cache / history for
     217             :      * the next frame or superframe.
     218             :      * @{
     219             :      */
     220             :     double prev_lsps[MAX_LSPS];   ///< LSPs of the last frame of the previous
     221             :                                   ///< superframe
     222             :     int last_pitch_val;           ///< pitch value of the previous frame
     223             :     int last_acb_type;            ///< frame type [0-2] of the previous frame
     224             :     int pitch_diff_sh16;          ///< ((cur_pitch_val - #last_pitch_val)
     225             :                                   ///< << 16) / #MAX_FRAMESIZE
     226             :     float silence_gain;           ///< set for use in blocks if #ACB_TYPE_NONE
     227             : 
     228             :     int aw_idx_is_ext;            ///< whether the AW index was encoded in
     229             :                                   ///< 8 bits (instead of 6)
     230             :     int aw_pulse_range;           ///< the range over which #aw_pulse_set1()
     231             :                                   ///< can apply the pulse, relative to the
     232             :                                   ///< value in aw_first_pulse_off. The exact
     233             :                                   ///< position of the first AW-pulse is within
     234             :                                   ///< [pulse_off, pulse_off + this], and
     235             :                                   ///< depends on bitstream values; [16 or 24]
     236             :     int aw_n_pulses[2];           ///< number of AW-pulses in each block; note
     237             :                                   ///< that this number can be negative (in
     238             :                                   ///< which case it basically means "zero")
     239             :     int aw_first_pulse_off[2];    ///< index of first sample to which to
     240             :                                   ///< apply AW-pulses, or -0xff if unset
     241             :     int aw_next_pulse_off_cache;  ///< the position (relative to start of the
     242             :                                   ///< second block) at which pulses should
     243             :                                   ///< start to be positioned, serves as a
     244             :                                   ///< cache for pitch-adaptive window pulses
     245             :                                   ///< between blocks
     246             : 
     247             :     int frame_cntr;               ///< current frame index [0 - 0xFFFE]; is
     248             :                                   ///< only used for comfort noise in #pRNG()
     249             :     int nb_superframes;           ///< number of superframes in current packet
     250             :     float gain_pred_err[6];       ///< cache for gain prediction
     251             :     float excitation_history[MAX_SIGNAL_HISTORY];
     252             :                                   ///< cache of the signal of previous
     253             :                                   ///< superframes, used as a history for
     254             :                                   ///< signal generation
     255             :     float synth_history[MAX_LSPS]; ///< see #excitation_history
     256             :     /**
     257             :      * @}
     258             :      *
     259             :      * @name Postfilter values
     260             :      *
     261             :      * Variables used for postfilter implementation, mostly history for
     262             :      * smoothing and so on, and context variables for FFT/iFFT.
     263             :      * @{
     264             :      */
     265             :     RDFTContext rdft, irdft;      ///< contexts for FFT-calculation in the
     266             :                                   ///< postfilter (for denoise filter)
     267             :     DCTContext dct, dst;          ///< contexts for phase shift (in Hilbert
     268             :                                   ///< transform, part of postfilter)
     269             :     float sin[511], cos[511];     ///< 8-bit cosine/sine windows over [-pi,pi]
     270             :                                   ///< range
     271             :     float postfilter_agc;         ///< gain control memory, used in
     272             :                                   ///< #adaptive_gain_control()
     273             :     float dcf_mem[2];             ///< DC filter history
     274             :     float zero_exc_pf[MAX_SIGNAL_HISTORY + MAX_SFRAMESIZE];
     275             :                                   ///< zero filter output (i.e. excitation)
     276             :                                   ///< by postfilter
     277             :     float denoise_filter_cache[MAX_FRAMESIZE];
     278             :     int   denoise_filter_cache_size; ///< samples in #denoise_filter_cache
     279             :     DECLARE_ALIGNED(32, float, tilted_lpcs_pf)[0x80];
     280             :                                   ///< aligned buffer for LPC tilting
     281             :     DECLARE_ALIGNED(32, float, denoise_coeffs_pf)[0x80];
     282             :                                   ///< aligned buffer for denoise coefficients
     283             :     DECLARE_ALIGNED(32, float, synth_filter_out_buf)[0x80 + MAX_LSPS_ALIGN16];
     284             :                                   ///< aligned buffer for postfilter speech
     285             :                                   ///< synthesis
     286             :     /**
     287             :      * @}
     288             :      */
     289             : } WMAVoiceContext;
     290             : 
     291             : /**
     292             :  * Set up the variable bit mode (VBM) tree from container extradata.
     293             :  * @param gb bit I/O context.
     294             :  *           The bit context (s->gb) should be loaded with byte 23-46 of the
     295             :  *           container extradata (i.e. the ones containing the VBM tree).
     296             :  * @param vbm_tree pointer to array to which the decoded VBM tree will be
     297             :  *                 written.
     298             :  * @return 0 on success, <0 on error.
     299             :  */
     300           8 : static av_cold int decode_vbmtree(GetBitContext *gb, int8_t vbm_tree[25])
     301             : {
     302           8 :     int cntr[8] = { 0 }, n, res;
     303             : 
     304           8 :     memset(vbm_tree, 0xff, sizeof(vbm_tree[0]) * 25);
     305         144 :     for (n = 0; n < 17; n++) {
     306         136 :         res = get_bits(gb, 3);
     307         136 :         if (cntr[res] > 3) // should be >= 3 + (res == 7))
     308           0 :             return -1;
     309         136 :         vbm_tree[res * 3 + cntr[res]++] = n;
     310             :     }
     311           8 :     return 0;
     312             : }
     313             : 
     314           5 : static av_cold void wmavoice_init_static_data(void)
     315             : {
     316             :     static const uint8_t bits[] = {
     317             :          2,  2,  2,  4,  4,  4,
     318             :          6,  6,  6,  8,  8,  8,
     319             :         10, 10, 10, 12, 12, 12,
     320             :         14, 14, 14, 14
     321             :     };
     322             :     static const uint16_t codes[] = {
     323             :           0x0000, 0x0001, 0x0002,        //              00/01/10
     324             :           0x000c, 0x000d, 0x000e,        //           11+00/01/10
     325             :           0x003c, 0x003d, 0x003e,        //         1111+00/01/10
     326             :           0x00fc, 0x00fd, 0x00fe,        //       111111+00/01/10
     327             :           0x03fc, 0x03fd, 0x03fe,        //     11111111+00/01/10
     328             :           0x0ffc, 0x0ffd, 0x0ffe,        //   1111111111+00/01/10
     329             :           0x3ffc, 0x3ffd, 0x3ffe, 0x3fff // 111111111111+xx
     330             :     };
     331             : 
     332           5 :     INIT_VLC_STATIC(&frame_type_vlc, VLC_NBITS, sizeof(bits),
     333             :                     bits, 1, 1, codes, 2, 2, 132);
     334           5 : }
     335             : 
     336           0 : static av_cold void wmavoice_flush(AVCodecContext *ctx)
     337             : {
     338           0 :     WMAVoiceContext *s = ctx->priv_data;
     339             :     int n;
     340             : 
     341           0 :     s->postfilter_agc    = 0;
     342           0 :     s->sframe_cache_size = 0;
     343           0 :     s->skip_bits_next    = 0;
     344           0 :     for (n = 0; n < s->lsps; n++)
     345           0 :         s->prev_lsps[n] = M_PI * (n + 1.0) / (s->lsps + 1.0);
     346           0 :     memset(s->excitation_history, 0,
     347             :            sizeof(*s->excitation_history) * MAX_SIGNAL_HISTORY);
     348           0 :     memset(s->synth_history,      0,
     349             :            sizeof(*s->synth_history)      * MAX_LSPS);
     350           0 :     memset(s->gain_pred_err,      0,
     351             :            sizeof(s->gain_pred_err));
     352             : 
     353           0 :     if (s->do_apf) {
     354           0 :         memset(&s->synth_filter_out_buf[MAX_LSPS_ALIGN16 - s->lsps], 0,
     355           0 :                sizeof(*s->synth_filter_out_buf) * s->lsps);
     356           0 :         memset(s->dcf_mem,              0,
     357             :                sizeof(*s->dcf_mem)              * 2);
     358           0 :         memset(s->zero_exc_pf,          0,
     359           0 :                sizeof(*s->zero_exc_pf)          * s->history_nsamples);
     360           0 :         memset(s->denoise_filter_cache, 0, sizeof(s->denoise_filter_cache));
     361             :     }
     362           0 : }
     363             : 
     364             : /**
     365             :  * Set up decoder with parameters from demuxer (extradata etc.).
     366             :  */
     367           8 : static av_cold int wmavoice_decode_init(AVCodecContext *ctx)
     368             : {
     369             :     static AVOnce init_static_once = AV_ONCE_INIT;
     370             :     int n, flags, pitch_range, lsp16_flag;
     371           8 :     WMAVoiceContext *s = ctx->priv_data;
     372             : 
     373           8 :     ff_thread_once(&init_static_once, wmavoice_init_static_data);
     374             : 
     375             :     /**
     376             :      * Extradata layout:
     377             :      * - byte  0-18: WMAPro-in-WMAVoice extradata (see wmaprodec.c),
     378             :      * - byte 19-22: flags field (annoyingly in LE; see below for known
     379             :      *               values),
     380             :      * - byte 23-46: variable bitmode tree (really just 17 * 3 bits,
     381             :      *               rest is 0).
     382             :      */
     383           8 :     if (ctx->extradata_size != 46) {
     384           0 :         av_log(ctx, AV_LOG_ERROR,
     385             :                "Invalid extradata size %d (should be 46)\n",
     386             :                ctx->extradata_size);
     387           0 :         return AVERROR_INVALIDDATA;
     388             :     }
     389           8 :     if (ctx->block_align <= 0) {
     390           0 :         av_log(ctx, AV_LOG_ERROR, "Invalid block alignment %d.\n", ctx->block_align);
     391           0 :         return AVERROR_INVALIDDATA;
     392             :     }
     393             : 
     394           8 :     flags                = AV_RL32(ctx->extradata + 18);
     395           8 :     s->spillover_bitsize = 3 + av_ceil_log2(ctx->block_align);
     396           8 :     s->do_apf            =    flags & 0x1;
     397           8 :     if (s->do_apf) {
     398           8 :         ff_rdft_init(&s->rdft,  7, DFT_R2C);
     399           8 :         ff_rdft_init(&s->irdft, 7, IDFT_C2R);
     400           8 :         ff_dct_init(&s->dct,  6, DCT_I);
     401           8 :         ff_dct_init(&s->dst,  6, DST_I);
     402             : 
     403           8 :         ff_sine_window_init(s->cos, 256);
     404           8 :         memcpy(&s->sin[255], s->cos, 256 * sizeof(s->cos[0]));
     405        2048 :         for (n = 0; n < 255; n++) {
     406        2040 :             s->sin[n]       = -s->sin[510 - n];
     407        2040 :             s->cos[510 - n] =  s->cos[n];
     408             :         }
     409             :     }
     410           8 :     s->denoise_strength  =   (flags >> 2) & 0xF;
     411           8 :     if (s->denoise_strength >= 12) {
     412           0 :         av_log(ctx, AV_LOG_ERROR,
     413             :                "Invalid denoise filter strength %d (max=11)\n",
     414             :                s->denoise_strength);
     415           0 :         return AVERROR_INVALIDDATA;
     416             :     }
     417           8 :     s->denoise_tilt_corr = !!(flags & 0x40);
     418           8 :     s->dc_level          =   (flags >> 7) & 0xF;
     419           8 :     s->lsp_q_mode        = !!(flags & 0x2000);
     420           8 :     s->lsp_def_mode      = !!(flags & 0x4000);
     421           8 :     lsp16_flag           =    flags & 0x1000;
     422           8 :     if (lsp16_flag) {
     423           4 :         s->lsps               = 16;
     424             :     } else {
     425           4 :         s->lsps               = 10;
     426             :     }
     427         112 :     for (n = 0; n < s->lsps; n++)
     428         104 :         s->prev_lsps[n] = M_PI * (n + 1.0) / (s->lsps + 1.0);
     429             : 
     430           8 :     init_get_bits(&s->gb, ctx->extradata + 22, (ctx->extradata_size - 22) << 3);
     431           8 :     if (decode_vbmtree(&s->gb, s->vbm_tree) < 0) {
     432           0 :         av_log(ctx, AV_LOG_ERROR, "Invalid VBM tree; broken extradata?\n");
     433           0 :         return AVERROR_INVALIDDATA;
     434             :     }
     435             : 
     436           8 :     s->min_pitch_val    = ((ctx->sample_rate << 8)      /  400 + 50) >> 8;
     437           8 :     s->max_pitch_val    = ((ctx->sample_rate << 8) * 37 / 2000 + 50) >> 8;
     438           8 :     pitch_range         = s->max_pitch_val - s->min_pitch_val;
     439           8 :     if (pitch_range <= 0) {
     440           0 :         av_log(ctx, AV_LOG_ERROR, "Invalid pitch range; broken extradata?\n");
     441           0 :         return AVERROR_INVALIDDATA;
     442             :     }
     443           8 :     s->pitch_nbits      = av_ceil_log2(pitch_range);
     444           8 :     s->last_pitch_val   = 40;
     445           8 :     s->last_acb_type    = ACB_TYPE_NONE;
     446           8 :     s->history_nsamples = s->max_pitch_val + 8;
     447             : 
     448           8 :     if (s->min_pitch_val < 1 || s->history_nsamples > MAX_SIGNAL_HISTORY) {
     449           0 :         int min_sr = ((((1 << 8) - 50) * 400) + 0xFF) >> 8,
     450           0 :             max_sr = ((((MAX_SIGNAL_HISTORY - 8) << 8) + 205) * 2000 / 37) >> 8;
     451             : 
     452           0 :         av_log(ctx, AV_LOG_ERROR,
     453             :                "Unsupported samplerate %d (min=%d, max=%d)\n",
     454             :                ctx->sample_rate, min_sr, max_sr); // 322-22097 Hz
     455             : 
     456           0 :         return AVERROR(ENOSYS);
     457             :     }
     458             : 
     459           8 :     s->block_conv_table[0]      = s->min_pitch_val;
     460           8 :     s->block_conv_table[1]      = (pitch_range * 25) >> 6;
     461           8 :     s->block_conv_table[2]      = (pitch_range * 44) >> 6;
     462           8 :     s->block_conv_table[3]      = s->max_pitch_val - 1;
     463           8 :     s->block_delta_pitch_hrange = (pitch_range >> 3) & ~0xF;
     464           8 :     if (s->block_delta_pitch_hrange <= 0) {
     465           0 :         av_log(ctx, AV_LOG_ERROR, "Invalid delta pitch hrange; broken extradata?\n");
     466           0 :         return AVERROR_INVALIDDATA;
     467             :     }
     468           8 :     s->block_delta_pitch_nbits  = 1 + av_ceil_log2(s->block_delta_pitch_hrange);
     469          24 :     s->block_pitch_range        = s->block_conv_table[2] +
     470          24 :                                   s->block_conv_table[3] + 1 +
     471           8 :                                   2 * (s->block_conv_table[1] - 2 * s->min_pitch_val);
     472           8 :     s->block_pitch_nbits        = av_ceil_log2(s->block_pitch_range);
     473             : 
     474           8 :     ctx->channels               = 1;
     475           8 :     ctx->channel_layout         = AV_CH_LAYOUT_MONO;
     476           8 :     ctx->sample_fmt             = AV_SAMPLE_FMT_FLT;
     477             : 
     478           8 :     return 0;
     479             : }
     480             : 
     481             : /**
     482             :  * @name Postfilter functions
     483             :  * Postfilter functions (gain control, wiener denoise filter, DC filter,
     484             :  * kalman smoothening, plus surrounding code to wrap it)
     485             :  * @{
     486             :  */
     487             : /**
     488             :  * Adaptive gain control (as used in postfilter).
     489             :  *
     490             :  * Identical to #ff_adaptive_gain_control() in acelp_vectors.c, except
     491             :  * that the energy here is calculated using sum(abs(...)), whereas the
     492             :  * other codecs (e.g. AMR-NB, SIPRO) use sqrt(dotproduct(...)).
     493             :  *
     494             :  * @param out output buffer for filtered samples
     495             :  * @param in input buffer containing the samples as they are after the
     496             :  *           postfilter steps so far
     497             :  * @param speech_synth input buffer containing speech synth before postfilter
     498             :  * @param size input buffer size
     499             :  * @param alpha exponential filter factor
     500             :  * @param gain_mem pointer to filter memory (single float)
     501             :  */
     502        6612 : static void adaptive_gain_control(float *out, const float *in,
     503             :                                   const float *speech_synth,
     504             :                                   int size, float alpha, float *gain_mem)
     505             : {
     506             :     int i;
     507        6612 :     float speech_energy = 0.0, postfilter_energy = 0.0, gain_scale_factor;
     508        6612 :     float mem = *gain_mem;
     509             : 
     510      535572 :     for (i = 0; i < size; i++) {
     511      528960 :         speech_energy     += fabsf(speech_synth[i]);
     512      528960 :         postfilter_energy += fabsf(in[i]);
     513             :     }
     514       13224 :     gain_scale_factor = postfilter_energy == 0.0 ? 0.0 :
     515        6612 :                         (1.0 - alpha) * speech_energy / postfilter_energy;
     516             : 
     517      535572 :     for (i = 0; i < size; i++) {
     518      528960 :         mem = alpha * mem + gain_scale_factor;
     519      528960 :         out[i] = in[i] * mem;
     520             :     }
     521             : 
     522        6612 :     *gain_mem = mem;
     523        6612 : }
     524             : 
     525             : /**
     526             :  * Kalman smoothing function.
     527             :  *
     528             :  * This function looks back pitch +/- 3 samples back into history to find
     529             :  * the best fitting curve (that one giving the optimal gain of the two
     530             :  * signals, i.e. the highest dot product between the two), and then
     531             :  * uses that signal history to smoothen the output of the speech synthesis
     532             :  * filter.
     533             :  *
     534             :  * @param s WMA Voice decoding context
     535             :  * @param pitch pitch of the speech signal
     536             :  * @param in input speech signal
     537             :  * @param out output pointer for smoothened signal
     538             :  * @param size input/output buffer size
     539             :  *
     540             :  * @returns -1 if no smoothening took place, e.g. because no optimal
     541             :  *          fit could be found, or 0 on success.
     542             :  */
     543        5070 : static int kalman_smoothen(WMAVoiceContext *s, int pitch,
     544             :                            const float *in, float *out, int size)
     545             : {
     546             :     int n;
     547        5070 :     float optimal_gain = 0, dot;
     548        5070 :     const float *ptr = &in[-FFMAX(s->min_pitch_val, pitch - 3)],
     549        5070 :                 *end = &in[-FFMIN(s->max_pitch_val, pitch + 3)],
     550        5070 :                 *best_hist_ptr = NULL;
     551             : 
     552             :     /* find best fitting point in history */
     553             :     do {
     554       35388 :         dot = avpriv_scalarproduct_float_c(in, ptr, size);
     555       35388 :         if (dot > optimal_gain) {
     556       12328 :             optimal_gain  = dot;
     557       12328 :             best_hist_ptr = ptr;
     558             :         }
     559       35388 :     } while (--ptr >= end);
     560             : 
     561        5070 :     if (optimal_gain <= 0)
     562          26 :         return -1;
     563        5044 :     dot = avpriv_scalarproduct_float_c(best_hist_ptr, best_hist_ptr, size);
     564        5044 :     if (dot <= 0) // would be 1.0
     565           0 :         return -1;
     566             : 
     567        5044 :     if (optimal_gain <= dot) {
     568        4872 :         dot = dot / (dot + 0.6 * optimal_gain); // 0.625-1.000
     569             :     } else
     570         172 :         dot = 0.625;
     571             : 
     572             :     /* actual smoothing */
     573      408564 :     for (n = 0; n < size; n++)
     574      403520 :         out[n] = best_hist_ptr[n] + dot * (in[n] - best_hist_ptr[n]);
     575             : 
     576        5044 :     return 0;
     577             : }
     578             : 
     579             : /**
     580             :  * Get the tilt factor of a formant filter from its transfer function
     581             :  * @see #tilt_factor() in amrnbdec.c, which does essentially the same,
     582             :  *      but somehow (??) it does a speech synthesis filter in the
     583             :  *      middle, which is missing here
     584             :  *
     585             :  * @param lpcs LPC coefficients
     586             :  * @param n_lpcs Size of LPC buffer
     587             :  * @returns the tilt factor
     588             :  */
     589        7098 : static float tilt_factor(const float *lpcs, int n_lpcs)
     590             : {
     591             :     float rh0, rh1;
     592             : 
     593        7098 :     rh0 = 1.0     + avpriv_scalarproduct_float_c(lpcs,  lpcs,    n_lpcs);
     594        7098 :     rh1 = lpcs[0] + avpriv_scalarproduct_float_c(lpcs, &lpcs[1], n_lpcs - 1);
     595             : 
     596        7098 :     return rh1 / rh0;
     597             : }
     598             : 
     599             : /**
     600             :  * Derive denoise filter coefficients (in real domain) from the LPCs.
     601             :  */
     602        5614 : static void calc_input_response(WMAVoiceContext *s, float *lpcs,
     603             :                                 int fcb_type, float *coeffs, int remainder)
     604             : {
     605        5614 :     float last_coeff, min = 15.0, max = -15.0;
     606             :     float irange, angle_mul, gain_mul, range, sq;
     607             :     int n, idx;
     608             : 
     609             :     /* Create frequency power spectrum of speech input (i.e. RDFT of LPCs) */
     610        5614 :     s->rdft.rdft_calc(&s->rdft, lpcs);
     611             : #define log_range(var, assign) do { \
     612             :         float tmp = log10f(assign);  var = tmp; \
     613             :         max       = FFMAX(max, tmp); min = FFMIN(min, tmp); \
     614             :     } while (0)
     615        5614 :     log_range(last_coeff,  lpcs[1]         * lpcs[1]);
     616      359296 :     for (n = 1; n < 64; n++)
     617      353682 :         log_range(lpcs[n], lpcs[n * 2]     * lpcs[n * 2] +
     618             :                            lpcs[n * 2 + 1] * lpcs[n * 2 + 1]);
     619        5614 :     log_range(lpcs[0],     lpcs[0]         * lpcs[0]);
     620             : #undef log_range
     621        5614 :     range    = max - min;
     622        5614 :     lpcs[64] = last_coeff;
     623             : 
     624             :     /* Now, use this spectrum to pick out these frequencies with higher
     625             :      * (relative) power/energy (which we then take to be "not noise"),
     626             :      * and set up a table (still in lpc[]) of (relative) gains per frequency.
     627             :      * These frequencies will be maintained, while others ("noise") will be
     628             :      * decreased in the filter output. */
     629        5614 :     irange    = 64.0 / range; // so irange*(max-value) is in the range [0, 63]
     630        5614 :     gain_mul  = range * (fcb_type == FCB_TYPE_HARDCODED ? (5.0 / 13.0) :
     631             :                                                           (5.0 / 14.7));
     632        5614 :     angle_mul = gain_mul * (8.0 * M_LN10 / M_PI);
     633      370524 :     for (n = 0; n <= 64; n++) {
     634             :         float pwr;
     635             : 
     636      364910 :         idx = FFMAX(0, lrint((max - lpcs[n]) * irange) - 1);
     637      364910 :         pwr = wmavoice_denoise_power_table[s->denoise_strength][idx];
     638      364910 :         lpcs[n] = angle_mul * pwr;
     639             : 
     640             :         /* 70.57 =~ 1/log10(1.0331663) */
     641      364910 :         idx = (pwr * gain_mul - 0.0295) * 70.570526123;
     642      364910 :         if (idx > 127) { // fall back if index falls outside table range
     643       17114 :             coeffs[n] = wmavoice_energy_table[127] *
     644        8557 :                         powf(1.0331663, idx - 127);
     645             :         } else
     646      356353 :             coeffs[n] = wmavoice_energy_table[FFMAX(0, idx)];
     647             :     }
     648             : 
     649             :     /* calculate the Hilbert transform of the gains, which we do (since this
     650             :      * is a sine input) by doing a phase shift (in theory, H(sin())=cos()).
     651             :      * Hilbert_Transform(RDFT(x)) = Laplace_Transform(x), which calculates the
     652             :      * "moment" of the LPCs in this filter. */
     653        5614 :     s->dct.dct_calc(&s->dct, lpcs);
     654        5614 :     s->dst.dct_calc(&s->dst, lpcs);
     655             : 
     656             :     /* Split out the coefficient indexes into phase/magnitude pairs */
     657        5614 :     idx = 255 + av_clip(lpcs[64],               -255, 255);
     658        5614 :     coeffs[0]  = coeffs[0]  * s->cos[idx];
     659        5614 :     idx = 255 + av_clip(lpcs[64] - 2 * lpcs[63], -255, 255);
     660        5614 :     last_coeff = coeffs[64] * s->cos[idx];
     661      179648 :     for (n = 63;; n--) {
     662      353682 :         idx = 255 + av_clip(-lpcs[64] - 2 * lpcs[n - 1], -255, 255);
     663      179648 :         coeffs[n * 2 + 1] = coeffs[n] * s->sin[idx];
     664      179648 :         coeffs[n * 2]     = coeffs[n] * s->cos[idx];
     665             : 
     666      179648 :         if (!--n) break;
     667             : 
     668      174034 :         idx = 255 + av_clip( lpcs[64] - 2 * lpcs[n - 1], -255, 255);
     669      174034 :         coeffs[n * 2 + 1] = coeffs[n] * s->sin[idx];
     670      174034 :         coeffs[n * 2]     = coeffs[n] * s->cos[idx];
     671             :     }
     672        5614 :     coeffs[1] = last_coeff;
     673             : 
     674             :     /* move into real domain */
     675        5614 :     s->irdft.rdft_calc(&s->irdft, coeffs);
     676             : 
     677             :     /* tilt correction and normalize scale */
     678        5614 :     memset(&coeffs[remainder], 0, sizeof(coeffs[0]) * (128 - remainder));
     679        5614 :     if (s->denoise_tilt_corr) {
     680        1484 :         float tilt_mem = 0;
     681             : 
     682        1484 :         coeffs[remainder - 1] = 0;
     683        1484 :         ff_tilt_compensation(&tilt_mem,
     684        1484 :                              -1.8 * tilt_factor(coeffs, remainder - 1),
     685             :                              coeffs, remainder);
     686             :     }
     687        5614 :     sq = (1.0 / 64.0) * sqrtf(1 / avpriv_scalarproduct_float_c(coeffs, coeffs,
     688             :                                                                remainder));
     689      269472 :     for (n = 0; n < remainder; n++)
     690      263858 :         coeffs[n] *= sq;
     691        5614 : }
     692             : 
     693             : /**
     694             :  * This function applies a Wiener filter on the (noisy) speech signal as
     695             :  * a means to denoise it.
     696             :  *
     697             :  * - take RDFT of LPCs to get the power spectrum of the noise + speech;
     698             :  * - using this power spectrum, calculate (for each frequency) the Wiener
     699             :  *    filter gain, which depends on the frequency power and desired level
     700             :  *    of noise subtraction (when set too high, this leads to artifacts)
     701             :  *    We can do this symmetrically over the X-axis (so 0-4kHz is the inverse
     702             :  *    of 4-8kHz);
     703             :  * - by doing a phase shift, calculate the Hilbert transform of this array
     704             :  *    of per-frequency filter-gains to get the filtering coefficients;
     705             :  * - smoothen/normalize/de-tilt these filter coefficients as desired;
     706             :  * - take RDFT of noisy sound, apply the coefficients and take its IRDFT
     707             :  *    to get the denoised speech signal;
     708             :  * - the leftover (i.e. output of the IRDFT on denoised speech data beyond
     709             :  *    the frame boundary) are saved and applied to subsequent frames by an
     710             :  *    overlap-add method (otherwise you get clicking-artifacts).
     711             :  *
     712             :  * @param s WMA Voice decoding context
     713             :  * @param fcb_type Frame (codebook) type
     714             :  * @param synth_pf input: the noisy speech signal, output: denoised speech
     715             :  *                 data; should be 16-byte aligned (for ASM purposes)
     716             :  * @param size size of the speech data
     717             :  * @param lpcs LPCs used to synthesize this frame's speech data
     718             :  */
     719        6612 : static void wiener_denoise(WMAVoiceContext *s, int fcb_type,
     720             :                            float *synth_pf, int size,
     721             :                            const float *lpcs)
     722             : {
     723             :     int remainder, lim, n;
     724             : 
     725        6612 :     if (fcb_type != FCB_TYPE_SILENCE) {
     726        5614 :         float *tilted_lpcs = s->tilted_lpcs_pf,
     727        5614 :               *coeffs = s->denoise_coeffs_pf, tilt_mem = 0;
     728             : 
     729        5614 :         tilted_lpcs[0]           = 1.0;
     730        5614 :         memcpy(&tilted_lpcs[1], lpcs, sizeof(lpcs[0]) * s->lsps);
     731        5614 :         memset(&tilted_lpcs[s->lsps + 1], 0,
     732        5614 :                sizeof(tilted_lpcs[0]) * (128 - s->lsps - 1));
     733        5614 :         ff_tilt_compensation(&tilt_mem, 0.7 * tilt_factor(lpcs, s->lsps),
     734        5614 :                              tilted_lpcs, s->lsps + 2);
     735             : 
     736             :         /* The IRDFT output (127 samples for 7-bit filter) beyond the frame
     737             :          * size is applied to the next frame. All input beyond this is zero,
     738             :          * and thus all output beyond this will go towards zero, hence we can
     739             :          * limit to min(size-1, 127-size) as a performance consideration. */
     740        5614 :         remainder = FFMIN(127 - size, size - 1);
     741        5614 :         calc_input_response(s, tilted_lpcs, fcb_type, coeffs, remainder);
     742             : 
     743             :         /* apply coefficients (in frequency spectrum domain), i.e. complex
     744             :          * number multiplication */
     745        5614 :         memset(&synth_pf[size], 0, sizeof(synth_pf[0]) * (128 - size));
     746        5614 :         s->rdft.rdft_calc(&s->rdft, synth_pf);
     747        5614 :         s->rdft.rdft_calc(&s->rdft, coeffs);
     748        5614 :         synth_pf[0] *= coeffs[0];
     749        5614 :         synth_pf[1] *= coeffs[1];
     750      359296 :         for (n = 1; n < 64; n++) {
     751      353682 :             float v1 = synth_pf[n * 2], v2 = synth_pf[n * 2 + 1];
     752      353682 :             synth_pf[n * 2]     = v1 * coeffs[n * 2] - v2 * coeffs[n * 2 + 1];
     753      353682 :             synth_pf[n * 2 + 1] = v2 * coeffs[n * 2] + v1 * coeffs[n * 2 + 1];
     754             :         }
     755        5614 :         s->irdft.rdft_calc(&s->irdft, synth_pf);
     756             :     }
     757             : 
     758             :     /* merge filter output with the history of previous runs */
     759        6612 :     if (s->denoise_filter_cache_size) {
     760        5612 :         lim = FFMIN(s->denoise_filter_cache_size, size);
     761      269376 :         for (n = 0; n < lim; n++)
     762      263764 :             synth_pf[n] += s->denoise_filter_cache[n];
     763        5612 :         s->denoise_filter_cache_size -= lim;
     764        5612 :         memmove(s->denoise_filter_cache, &s->denoise_filter_cache[size],
     765        5612 :                 sizeof(s->denoise_filter_cache[0]) * s->denoise_filter_cache_size);
     766             :     }
     767             : 
     768             :     /* move remainder of filter output into a cache for future runs */
     769        6612 :     if (fcb_type != FCB_TYPE_SILENCE) {
     770        5614 :         lim = FFMIN(remainder, s->denoise_filter_cache_size);
     771        5614 :         for (n = 0; n < lim; n++)
     772           0 :             s->denoise_filter_cache[n] += synth_pf[size + n];
     773        5614 :         if (lim < remainder) {
     774        5614 :             memcpy(&s->denoise_filter_cache[lim], &synth_pf[size + lim],
     775        5614 :                    sizeof(s->denoise_filter_cache[0]) * (remainder - lim));
     776        5614 :             s->denoise_filter_cache_size = remainder;
     777             :         }
     778             :     }
     779        6612 : }
     780             : 
     781             : /**
     782             :  * Averaging projection filter, the postfilter used in WMAVoice.
     783             :  *
     784             :  * This uses the following steps:
     785             :  * - A zero-synthesis filter (generate excitation from synth signal)
     786             :  * - Kalman smoothing on excitation, based on pitch
     787             :  * - Re-synthesized smoothened output
     788             :  * - Iterative Wiener denoise filter
     789             :  * - Adaptive gain filter
     790             :  * - DC filter
     791             :  *
     792             :  * @param s WMAVoice decoding context
     793             :  * @param synth Speech synthesis output (before postfilter)
     794             :  * @param samples Output buffer for filtered samples
     795             :  * @param size Buffer size of synth & samples
     796             :  * @param lpcs Generated LPCs used for speech synthesis
     797             :  * @param zero_exc_pf destination for zero synthesis filter (16-byte aligned)
     798             :  * @param fcb_type Frame type (silence, hardcoded, AW-pulses or FCB-pulses)
     799             :  * @param pitch Pitch of the input signal
     800             :  */
     801        6612 : static void postfilter(WMAVoiceContext *s, const float *synth,
     802             :                        float *samples,    int size,
     803             :                        const float *lpcs, float *zero_exc_pf,
     804             :                        int fcb_type,      int pitch)
     805             : {
     806             :     float synth_filter_in_buf[MAX_FRAMESIZE / 2],
     807        6612 :           *synth_pf = &s->synth_filter_out_buf[MAX_LSPS_ALIGN16],
     808        6612 :           *synth_filter_in = zero_exc_pf;
     809             : 
     810        6612 :     av_assert0(size <= MAX_FRAMESIZE / 2);
     811             : 
     812             :     /* generate excitation from input signal */
     813        6612 :     ff_celp_lp_zero_synthesis_filterf(zero_exc_pf, lpcs, synth, size, s->lsps);
     814             : 
     815       11682 :     if (fcb_type >= FCB_TYPE_AW_PULSES &&
     816        5070 :         !kalman_smoothen(s, pitch, zero_exc_pf, synth_filter_in_buf, size))
     817        5044 :         synth_filter_in = synth_filter_in_buf;
     818             : 
     819             :     /* re-synthesize speech after smoothening, and keep history */
     820        6612 :     ff_celp_lp_synthesis_filterf(synth_pf, lpcs,
     821             :                                  synth_filter_in, size, s->lsps);
     822        6612 :     memcpy(&synth_pf[-s->lsps], &synth_pf[size - s->lsps],
     823        6612 :            sizeof(synth_pf[0]) * s->lsps);
     824             : 
     825        6612 :     wiener_denoise(s, fcb_type, synth_pf, size, lpcs);
     826             : 
     827        6612 :     adaptive_gain_control(samples, synth_pf, synth, size, 0.99,
     828             :                           &s->postfilter_agc);
     829             : 
     830        6612 :     if (s->dc_level > 8) {
     831             :         /* remove ultra-low frequency DC noise / highpass filter;
     832             :          * coefficients are identical to those used in SIPR decoding,
     833             :          * and very closely resemble those used in AMR-NB decoding. */
     834           0 :         ff_acelp_apply_order_2_transfer_function(samples, samples,
     835           0 :             (const float[2]) { -1.99997,      1.0 },
     836           0 :             (const float[2]) { -1.9330735188, 0.93589198496 },
     837           0 :             0.93980580475, s->dcf_mem, size);
     838             :     }
     839        6612 : }
     840             : /**
     841             :  * @}
     842             :  */
     843             : 
     844             : /**
     845             :  * Dequantize LSPs
     846             :  * @param lsps output pointer to the array that will hold the LSPs
     847             :  * @param num number of LSPs to be dequantized
     848             :  * @param values quantized values, contains n_stages values
     849             :  * @param sizes range (i.e. max value) of each quantized value
     850             :  * @param n_stages number of dequantization runs
     851             :  * @param table dequantization table to be used
     852             :  * @param mul_q LSF multiplier
     853             :  * @param base_q base (lowest) LSF values
     854             :  */
     855        4404 : static void dequant_lsps(double *lsps, int num,
     856             :                          const uint16_t *values,
     857             :                          const uint16_t *sizes,
     858             :                          int n_stages, const uint8_t *table,
     859             :                          const double *mul_q,
     860             :                          const double *base_q)
     861             : {
     862             :     int n, m;
     863             : 
     864        4404 :     memset(lsps, 0, num * sizeof(*lsps));
     865       12668 :     for (n = 0; n < n_stages; n++) {
     866        8264 :         const uint8_t *t_off = &table[values[n] * num];
     867        8264 :         double base = base_q[n], mul = mul_q[n];
     868             : 
     869       95364 :         for (m = 0; m < num; m++)
     870       87100 :             lsps[m] += base + mul * t_off[m];
     871             : 
     872        8264 :         table += sizes[n] * num;
     873             :     }
     874        4404 : }
     875             : 
     876             : /**
     877             :  * @name LSP dequantization routines
     878             :  * LSP dequantization routines, for 10/16LSPs and independent/residual coding.
     879             :  * lsp10i() consumes 24 bits; lsp10r() consumes an additional 24 bits;
     880             :  * lsp16i() consumes 34 bits; lsp16r() consumes an additional 26 bits.
     881             :  * @{
     882             :  */
     883             : /**
     884             :  * Parse 10 independently-coded LSPs.
     885             :  */
     886         552 : static void dequant_lsp10i(GetBitContext *gb, double *lsps)
     887             : {
     888             :     static const uint16_t vec_sizes[4] = { 256, 64, 32, 32 };
     889             :     static const double mul_lsf[4] = {
     890             :         5.2187144800e-3,    1.4626986422e-3,
     891             :         9.6179549166e-4,    1.1325736225e-3
     892             :     };
     893             :     static const double base_lsf[4] = {
     894             :         M_PI * -2.15522e-1, M_PI * -6.1646e-2,
     895             :         M_PI * -3.3486e-2,  M_PI * -5.7408e-2
     896             :     };
     897             :     uint16_t v[4];
     898             : 
     899         552 :     v[0] = get_bits(gb, 8);
     900         552 :     v[1] = get_bits(gb, 6);
     901         552 :     v[2] = get_bits(gb, 5);
     902         552 :     v[3] = get_bits(gb, 5);
     903             : 
     904         552 :     dequant_lsps(lsps, 10, v, vec_sizes, 4, wmavoice_dq_lsp10i,
     905             :                  mul_lsf, base_lsf);
     906         552 : }
     907             : 
     908             : /**
     909             :  * Parse 10 independently-coded LSPs, and then derive the tables to
     910             :  * generate LSPs for the other frames from them (residual coding).
     911             :  */
     912         552 : static void dequant_lsp10r(GetBitContext *gb,
     913             :                            double *i_lsps, const double *old,
     914             :                            double *a1, double *a2, int q_mode)
     915             : {
     916             :     static const uint16_t vec_sizes[3] = { 128, 64, 64 };
     917             :     static const double mul_lsf[3] = {
     918             :         2.5807601174e-3,    1.2354460219e-3,   1.1763821673e-3
     919             :     };
     920             :     static const double base_lsf[3] = {
     921             :         M_PI * -1.07448e-1, M_PI * -5.2706e-2, M_PI * -5.1634e-2
     922             :     };
     923         552 :     const float (*ipol_tab)[2][10] = q_mode ?
     924         552 :         wmavoice_lsp10_intercoeff_b : wmavoice_lsp10_intercoeff_a;
     925             :     uint16_t interpol, v[3];
     926             :     int n;
     927             : 
     928         552 :     dequant_lsp10i(gb, i_lsps);
     929             : 
     930         552 :     interpol = get_bits(gb, 5);
     931         552 :     v[0]     = get_bits(gb, 7);
     932         552 :     v[1]     = get_bits(gb, 6);
     933         552 :     v[2]     = get_bits(gb, 6);
     934             : 
     935        6072 :     for (n = 0; n < 10; n++) {
     936        5520 :         double delta = old[n] - i_lsps[n];
     937        5520 :         a1[n]        = ipol_tab[interpol][0][n] * delta + i_lsps[n];
     938        5520 :         a1[10 + n]   = ipol_tab[interpol][1][n] * delta + i_lsps[n];
     939             :     }
     940             : 
     941         552 :     dequant_lsps(a2, 20, v, vec_sizes, 3, wmavoice_dq_lsp10r,
     942             :                  mul_lsf, base_lsf);
     943         552 : }
     944             : 
     945             : /**
     946             :  * Parse 16 independently-coded LSPs.
     947             :  */
     948         550 : static void dequant_lsp16i(GetBitContext *gb, double *lsps)
     949             : {
     950             :     static const uint16_t vec_sizes[5] = { 256, 64, 128, 64, 128 };
     951             :     static const double mul_lsf[5] = {
     952             :         3.3439586280e-3,    6.9908173703e-4,
     953             :         3.3216608306e-3,    1.0334960326e-3,
     954             :         3.1899104283e-3
     955             :     };
     956             :     static const double base_lsf[5] = {
     957             :         M_PI * -1.27576e-1, M_PI * -2.4292e-2,
     958             :         M_PI * -1.28094e-1, M_PI * -3.2128e-2,
     959             :         M_PI * -1.29816e-1
     960             :     };
     961             :     uint16_t v[5];
     962             : 
     963         550 :     v[0] = get_bits(gb, 8);
     964         550 :     v[1] = get_bits(gb, 6);
     965         550 :     v[2] = get_bits(gb, 7);
     966         550 :     v[3] = get_bits(gb, 6);
     967         550 :     v[4] = get_bits(gb, 7);
     968             : 
     969         550 :     dequant_lsps( lsps,     5,  v,     vec_sizes,    2,
     970             :                  wmavoice_dq_lsp16i1,  mul_lsf,     base_lsf);
     971         550 :     dequant_lsps(&lsps[5],  5, &v[2], &vec_sizes[2], 2,
     972             :                  wmavoice_dq_lsp16i2, &mul_lsf[2], &base_lsf[2]);
     973         550 :     dequant_lsps(&lsps[10], 6, &v[4], &vec_sizes[4], 1,
     974             :                  wmavoice_dq_lsp16i3, &mul_lsf[4], &base_lsf[4]);
     975         550 : }
     976             : 
     977             : /**
     978             :  * Parse 16 independently-coded LSPs, and then derive the tables to
     979             :  * generate LSPs for the other frames from them (residual coding).
     980             :  */
     981         550 : static void dequant_lsp16r(GetBitContext *gb,
     982             :                            double *i_lsps, const double *old,
     983             :                            double *a1, double *a2, int q_mode)
     984             : {
     985             :     static const uint16_t vec_sizes[3] = { 128, 128, 128 };
     986             :     static const double mul_lsf[3] = {
     987             :         1.2232979501e-3,   1.4062241527e-3,   1.6114744851e-3
     988             :     };
     989             :     static const double base_lsf[3] = {
     990             :         M_PI * -5.5830e-2, M_PI * -5.2908e-2, M_PI * -5.4776e-2
     991             :     };
     992         550 :     const float (*ipol_tab)[2][16] = q_mode ?
     993         550 :         wmavoice_lsp16_intercoeff_b : wmavoice_lsp16_intercoeff_a;
     994             :     uint16_t interpol, v[3];
     995             :     int n;
     996             : 
     997         550 :     dequant_lsp16i(gb, i_lsps);
     998             : 
     999         550 :     interpol = get_bits(gb, 5);
    1000         550 :     v[0]     = get_bits(gb, 7);
    1001         550 :     v[1]     = get_bits(gb, 7);
    1002         550 :     v[2]     = get_bits(gb, 7);
    1003             : 
    1004        9350 :     for (n = 0; n < 16; n++) {
    1005        8800 :         double delta = old[n] - i_lsps[n];
    1006        8800 :         a1[n]        = ipol_tab[interpol][0][n] * delta + i_lsps[n];
    1007        8800 :         a1[16 + n]   = ipol_tab[interpol][1][n] * delta + i_lsps[n];
    1008             :     }
    1009             : 
    1010         550 :     dequant_lsps( a2,     10,  v,     vec_sizes,    1,
    1011             :                  wmavoice_dq_lsp16r1,  mul_lsf,     base_lsf);
    1012         550 :     dequant_lsps(&a2[10], 10, &v[1], &vec_sizes[1], 1,
    1013             :                  wmavoice_dq_lsp16r2, &mul_lsf[1], &base_lsf[1]);
    1014         550 :     dequant_lsps(&a2[20], 12, &v[2], &vec_sizes[2], 1,
    1015             :                  wmavoice_dq_lsp16r3, &mul_lsf[2], &base_lsf[2]);
    1016         550 : }
    1017             : 
    1018             : /**
    1019             :  * @}
    1020             :  * @name Pitch-adaptive window coding functions
    1021             :  * The next few functions are for pitch-adaptive window coding.
    1022             :  * @{
    1023             :  */
    1024             : /**
    1025             :  * Parse the offset of the first pitch-adaptive window pulses, and
    1026             :  * the distribution of pulses between the two blocks in this frame.
    1027             :  * @param s WMA Voice decoding context private data
    1028             :  * @param gb bit I/O context
    1029             :  * @param pitch pitch for each block in this frame
    1030             :  */
    1031         341 : static void aw_parse_coords(WMAVoiceContext *s, GetBitContext *gb,
    1032             :                             const int *pitch)
    1033             : {
    1034             :     static const int16_t start_offset[94] = {
    1035             :         -11,  -9,  -7,  -5,  -3,  -1,   1,   3,   5,   7,   9,  11,
    1036             :          13,  15,  18,  17,  19,  20,  21,  22,  23,  24,  25,  26,
    1037             :          27,  28,  29,  30,  31,  32,  33,  35,  37,  39,  41,  43,
    1038             :          45,  47,  49,  51,  53,  55,  57,  59,  61,  63,  65,  67,
    1039             :          69,  71,  73,  75,  77,  79,  81,  83,  85,  87,  89,  91,
    1040             :          93,  95,  97,  99, 101, 103, 105, 107, 109, 111, 113, 115,
    1041             :         117, 119, 121, 123, 125, 127, 129, 131, 133, 135, 137, 139,
    1042             :         141, 143, 145, 147, 149, 151, 153, 155, 157, 159
    1043             :     };
    1044             :     int bits, offset;
    1045             : 
    1046             :     /* position of pulse */
    1047         341 :     s->aw_idx_is_ext = 0;
    1048         341 :     if ((bits = get_bits(gb, 6)) >= 54) {
    1049          10 :         s->aw_idx_is_ext = 1;
    1050          10 :         bits += (bits - 54) * 3 + get_bits(gb, 2);
    1051             :     }
    1052             : 
    1053             :     /* for a repeated pulse at pulse_off with a pitch_lag of pitch[], count
    1054             :      * the distribution of the pulses in each block contained in this frame. */
    1055         341 :     s->aw_pulse_range        = FFMIN(pitch[0], pitch[1]) > 32 ? 24 : 16;
    1056         341 :     for (offset = start_offset[bits]; offset < 0; offset += pitch[0]) ;
    1057         341 :     s->aw_n_pulses[0]        = (pitch[0] - 1 + MAX_FRAMESIZE / 2 - offset) / pitch[0];
    1058         341 :     s->aw_first_pulse_off[0] = offset - s->aw_pulse_range / 2;
    1059         341 :     offset                  += s->aw_n_pulses[0] * pitch[0];
    1060         341 :     s->aw_n_pulses[1]        = (pitch[1] - 1 + MAX_FRAMESIZE - offset) / pitch[1];
    1061         341 :     s->aw_first_pulse_off[1] = offset - (MAX_FRAMESIZE + s->aw_pulse_range) / 2;
    1062             : 
    1063             :     /* if continuing from a position before the block, reset position to
    1064             :      * start of block (when corrected for the range over which it can be
    1065             :      * spread in aw_pulse_set1()). */
    1066         341 :     if (start_offset[bits] < MAX_FRAMESIZE / 2) {
    1067         718 :         while (s->aw_first_pulse_off[1] - pitch[1] + s->aw_pulse_range > 0)
    1068          56 :             s->aw_first_pulse_off[1] -= pitch[1];
    1069         331 :         if (start_offset[bits] < 0)
    1070         150 :             while (s->aw_first_pulse_off[0] - pitch[0] + s->aw_pulse_range > 0)
    1071          50 :                 s->aw_first_pulse_off[0] -= pitch[0];
    1072             :     }
    1073         341 : }
    1074             : 
    1075             : /**
    1076             :  * Apply second set of pitch-adaptive window pulses.
    1077             :  * @param s WMA Voice decoding context private data
    1078             :  * @param gb bit I/O context
    1079             :  * @param block_idx block index in frame [0, 1]
    1080             :  * @param fcb structure containing fixed codebook vector info
    1081             :  * @return -1 on error, 0 otherwise
    1082             :  */
    1083         682 : static int aw_pulse_set2(WMAVoiceContext *s, GetBitContext *gb,
    1084             :                          int block_idx, AMRFixed *fcb)
    1085             : {
    1086             :     uint16_t use_mask_mem[9]; // only 5 are used, rest is padding
    1087         682 :     uint16_t *use_mask = use_mask_mem + 2;
    1088             :     /* in this function, idx is the index in the 80-bit (+ padding) use_mask
    1089             :      * bit-array. Since use_mask consists of 16-bit values, the lower 4 bits
    1090             :      * of idx are the position of the bit within a particular item in the
    1091             :      * array (0 being the most significant bit, and 15 being the least
    1092             :      * significant bit), and the remainder (>> 4) is the index in the
    1093             :      * use_mask[]-array. This is faster and uses less memory than using a
    1094             :      * 80-byte/80-int array. */
    1095         682 :     int pulse_off = s->aw_first_pulse_off[block_idx],
    1096         682 :         pulse_start, n, idx, range, aidx, start_off = 0;
    1097             : 
    1098             :     /* set offset of first pulse to within this block */
    1099         682 :     if (s->aw_n_pulses[block_idx] > 0)
    1100        1314 :         while (pulse_off + s->aw_pulse_range < 1)
    1101           0 :             pulse_off += fcb->pitch_lag;
    1102             : 
    1103             :     /* find range per pulse */
    1104         682 :     if (s->aw_n_pulses[0] > 0) {
    1105         646 :         if (block_idx == 0) {
    1106         323 :             range = 32;
    1107             :         } else /* block_idx = 1 */ {
    1108         323 :             range = 8;
    1109         323 :             if (s->aw_n_pulses[block_idx] > 0)
    1110         316 :                 pulse_off = s->aw_next_pulse_off_cache;
    1111             :         }
    1112             :     } else
    1113          36 :         range = 16;
    1114         682 :     pulse_start = s->aw_n_pulses[block_idx] > 0 ? pulse_off - range / 2 : 0;
    1115             : 
    1116             :     /* aw_pulse_set1() already applies pulses around pulse_off (to be exactly,
    1117             :      * in the range of [pulse_off, pulse_off + s->aw_pulse_range], and thus
    1118             :      * we exclude that range from being pulsed again in this function. */
    1119         682 :     memset(&use_mask[-2], 0, 2 * sizeof(use_mask[0]));
    1120         682 :     memset( use_mask,   -1, 5 * sizeof(use_mask[0]));
    1121         682 :     memset(&use_mask[5], 0, 2 * sizeof(use_mask[0]));
    1122         682 :     if (s->aw_n_pulses[block_idx] > 0)
    1123        1568 :         for (idx = pulse_off; idx < MAX_FRAMESIZE / 2; idx += fcb->pitch_lag) {
    1124         911 :             int excl_range         = s->aw_pulse_range; // always 16 or 24
    1125         911 :             uint16_t *use_mask_ptr = &use_mask[idx >> 4];
    1126         911 :             int first_sh           = 16 - (idx & 15);
    1127         911 :             *use_mask_ptr++       &= 0xFFFFu << first_sh;
    1128         911 :             excl_range            -= first_sh;
    1129         911 :             if (excl_range >= 16) {
    1130         468 :                 *use_mask_ptr++    = 0;
    1131         468 :                 *use_mask_ptr     &= 0xFFFF >> (excl_range - 16);
    1132             :             } else
    1133         443 :                 *use_mask_ptr     &= 0xFFFF >> excl_range;
    1134             :         }
    1135             : 
    1136             :     /* find the 'aidx'th offset that is not excluded */
    1137         682 :     aidx = get_bits(gb, s->aw_n_pulses[0] > 0 ? 5 - 2 * block_idx : 4);
    1138       16825 :     for (n = 0; n <= aidx; pulse_start++) {
    1139       16143 :         for (idx = pulse_start; idx < 0; idx += fcb->pitch_lag) ;
    1140       16143 :         if (idx >= MAX_FRAMESIZE / 2) { // find from zero
    1141         538 :             if (use_mask[0])      idx = 0x0F;
    1142         123 :             else if (use_mask[1]) idx = 0x1F;
    1143          18 :             else if (use_mask[2]) idx = 0x2F;
    1144           0 :             else if (use_mask[3]) idx = 0x3F;
    1145           0 :             else if (use_mask[4]) idx = 0x4F;
    1146           0 :             else return -1;
    1147         538 :             idx -= av_log2_16bit(use_mask[idx >> 4]);
    1148             :         }
    1149       16143 :         if (use_mask[idx >> 4] & (0x8000 >> (idx & 15))) {
    1150        7465 :             use_mask[idx >> 4] &= ~(0x8000 >> (idx & 15));
    1151        7465 :             n++;
    1152        7465 :             start_off = idx;
    1153             :         }
    1154             :     }
    1155             : 
    1156         682 :     fcb->x[fcb->n] = start_off;
    1157         682 :     fcb->y[fcb->n] = get_bits1(gb) ? -1.0 : 1.0;
    1158         682 :     fcb->n++;
    1159             : 
    1160             :     /* set offset for next block, relative to start of that block */
    1161         682 :     n = (MAX_FRAMESIZE / 2 - start_off) % fcb->pitch_lag;
    1162         682 :     s->aw_next_pulse_off_cache = n ? fcb->pitch_lag - n : 0;
    1163         682 :     return 0;
    1164             : }
    1165             : 
    1166             : /**
    1167             :  * Apply first set of pitch-adaptive window pulses.
    1168             :  * @param s WMA Voice decoding context private data
    1169             :  * @param gb bit I/O context
    1170             :  * @param block_idx block index in frame [0, 1]
    1171             :  * @param fcb storage location for fixed codebook pulse info
    1172             :  */
    1173         682 : static void aw_pulse_set1(WMAVoiceContext *s, GetBitContext *gb,
    1174             :                           int block_idx, AMRFixed *fcb)
    1175             : {
    1176         682 :     int val = get_bits(gb, 12 - 2 * (s->aw_idx_is_ext && !block_idx));
    1177             :     float v;
    1178             : 
    1179         682 :     if (s->aw_n_pulses[block_idx] > 0) {
    1180             :         int n, v_mask, i_mask, sh, n_pulses;
    1181             : 
    1182         657 :         if (s->aw_pulse_range == 24) { // 3 pulses, 1:sign + 3:index each
    1183         652 :             n_pulses = 3;
    1184         652 :             v_mask   = 8;
    1185         652 :             i_mask   = 7;
    1186         652 :             sh       = 4;
    1187             :         } else { // 4 pulses, 1:sign + 2:index each
    1188           5 :             n_pulses = 4;
    1189           5 :             v_mask   = 4;
    1190           5 :             i_mask   = 3;
    1191           5 :             sh       = 3;
    1192             :         }
    1193             : 
    1194        2633 :         for (n = n_pulses - 1; n >= 0; n--, val >>= sh) {
    1195        1976 :             fcb->y[fcb->n] = (val & v_mask) ? -1.0 : 1.0;
    1196        3952 :             fcb->x[fcb->n] = (val & i_mask) * n_pulses + n +
    1197        1976 :                                  s->aw_first_pulse_off[block_idx];
    1198        4193 :             while (fcb->x[fcb->n] < 0)
    1199         241 :                 fcb->x[fcb->n] += fcb->pitch_lag;
    1200        1976 :             if (fcb->x[fcb->n] < MAX_FRAMESIZE / 2)
    1201        1959 :                 fcb->n++;
    1202             :         }
    1203             :     } else {
    1204          25 :         int num2 = (val & 0x1FF) >> 1, delta, idx;
    1205             : 
    1206          25 :         if (num2 < 1 * 79)      { delta = 1; idx = num2 + 1; }
    1207          21 :         else if (num2 < 2 * 78) { delta = 3; idx = num2 + 1 - 1 * 77; }
    1208          15 :         else if (num2 < 3 * 77) { delta = 5; idx = num2 + 1 - 2 * 76; }
    1209           5 :         else                    { delta = 7; idx = num2 + 1 - 3 * 75; }
    1210          25 :         v = (val & 0x200) ? -1.0 : 1.0;
    1211             : 
    1212          25 :         fcb->no_repeat_mask |= 3 << fcb->n;
    1213          25 :         fcb->x[fcb->n]       = idx - delta;
    1214          25 :         fcb->y[fcb->n]       = v;
    1215          25 :         fcb->x[fcb->n + 1]   = idx;
    1216          25 :         fcb->y[fcb->n + 1]   = (val & 1) ? -v : v;
    1217          25 :         fcb->n              += 2;
    1218             :     }
    1219         682 : }
    1220             : 
    1221             : /**
    1222             :  * @}
    1223             :  *
    1224             :  * Generate a random number from frame_cntr and block_idx, which will live
    1225             :  * in the range [0, 1000 - block_size] (so it can be used as an index in a
    1226             :  * table of size 1000 of which you want to read block_size entries).
    1227             :  *
    1228             :  * @param frame_cntr current frame number
    1229             :  * @param block_num current block index
    1230             :  * @param block_size amount of entries we want to read from a table
    1231             :  *                   that has 1000 entries
    1232             :  * @return a (non-)random number in the [0, 1000 - block_size] range.
    1233             :  */
    1234         499 : static int pRNG(int frame_cntr, int block_num, int block_size)
    1235             : {
    1236             :     /* array to simplify the calculation of z:
    1237             :      * y = (x % 9) * 5 + 6;
    1238             :      * z = (49995 * x) / y;
    1239             :      * Since y only has 9 values, we can remove the division by using a
    1240             :      * LUT and using FASTDIV-style divisions. For each of the 9 values
    1241             :      * of y, we can rewrite z as:
    1242             :      * z = x * (49995 / y) + x * ((49995 % y) / y)
    1243             :      * In this table, each col represents one possible value of y, the
    1244             :      * first number is 49995 / y, and the second is the FASTDIV variant
    1245             :      * of 49995 % y / y. */
    1246             :     static const unsigned int div_tbl[9][2] = {
    1247             :         { 8332,  3 * 715827883U }, // y =  6
    1248             :         { 4545,  0 * 390451573U }, // y = 11
    1249             :         { 3124, 11 * 268435456U }, // y = 16
    1250             :         { 2380, 15 * 204522253U }, // y = 21
    1251             :         { 1922, 23 * 165191050U }, // y = 26
    1252             :         { 1612, 23 * 138547333U }, // y = 31
    1253             :         { 1388, 27 * 119304648U }, // y = 36
    1254             :         { 1219, 16 * 104755300U }, // y = 41
    1255             :         { 1086, 39 *  93368855U }  // y = 46
    1256             :     };
    1257         499 :     unsigned int z, y, x = MUL16(block_num, 1877) + frame_cntr;
    1258         499 :     if (x >= 0xFFFF) x -= 0xFFFF;   // max value of x is 8*1877+0xFFFE=0x13AA6,
    1259             :                                     // so this is effectively a modulo (%)
    1260         499 :     y = x - 9 * MULH(477218589, x); // x % 9
    1261         499 :     z = (uint16_t) (x * div_tbl[y][0] + UMULH(x, div_tbl[y][1]));
    1262             :                                     // z = x * 49995 / (y * 5 + 6)
    1263         499 :     return z % (1000 - block_size);
    1264             : }
    1265             : 
    1266             : /**
    1267             :  * Parse hardcoded signal for a single block.
    1268             :  * @note see #synth_block().
    1269             :  */
    1270        1043 : static void synth_block_hardcoded(WMAVoiceContext *s, GetBitContext *gb,
    1271             :                                  int block_idx, int size,
    1272             :                                  const struct frame_type_desc *frame_desc,
    1273             :                                  float *excitation)
    1274             : {
    1275             :     float gain;
    1276             :     int n, r_idx;
    1277             : 
    1278        1043 :     av_assert0(size <= MAX_FRAMESIZE);
    1279             : 
    1280             :     /* Set the offset from which we start reading wmavoice_std_codebook */
    1281        1043 :     if (frame_desc->fcb_type == FCB_TYPE_SILENCE) {
    1282         499 :         r_idx = pRNG(s->frame_cntr, block_idx, size);
    1283         499 :         gain  = s->silence_gain;
    1284             :     } else /* FCB_TYPE_HARDCODED */ {
    1285         544 :         r_idx = get_bits(gb, 8);
    1286         544 :         gain  = wmavoice_gain_universal[get_bits(gb, 6)];
    1287             :     }
    1288             : 
    1289             :     /* Clear gain prediction parameters */
    1290        1043 :     memset(s->gain_pred_err, 0, sizeof(s->gain_pred_err));
    1291             : 
    1292             :     /* Apply gain to hardcoded codebook and use that as excitation signal */
    1293      124403 :     for (n = 0; n < size; n++)
    1294      123360 :         excitation[n] = wmavoice_std_codebook[r_idx + n] * gain;
    1295        1043 : }
    1296             : 
    1297             : /**
    1298             :  * Parse FCB/ACB signal for a single block.
    1299             :  * @note see #synth_block().
    1300             :  */
    1301        9740 : static void synth_block_fcb_acb(WMAVoiceContext *s, GetBitContext *gb,
    1302             :                                 int block_idx, int size,
    1303             :                                 int block_pitch_sh2,
    1304             :                                 const struct frame_type_desc *frame_desc,
    1305             :                                 float *excitation)
    1306             : {
    1307             :     static const float gain_coeff[6] = {
    1308             :         0.8169, -0.06545, 0.1726, 0.0185, -0.0359, 0.0458
    1309             :     };
    1310             :     float pulses[MAX_FRAMESIZE / 2], pred_err, acb_gain, fcb_gain;
    1311             :     int n, idx, gain_weight;
    1312             :     AMRFixed fcb;
    1313             : 
    1314        9740 :     av_assert0(size <= MAX_FRAMESIZE / 2);
    1315        9740 :     memset(pulses, 0, sizeof(*pulses) * size);
    1316             : 
    1317        9740 :     fcb.pitch_lag      = block_pitch_sh2 >> 2;
    1318        9740 :     fcb.pitch_fac      = 1.0;
    1319        9740 :     fcb.no_repeat_mask = 0;
    1320        9740 :     fcb.n              = 0;
    1321             : 
    1322             :     /* For the other frame types, this is where we apply the innovation
    1323             :      * (fixed) codebook pulses of the speech signal. */
    1324        9740 :     if (frame_desc->fcb_type == FCB_TYPE_AW_PULSES) {
    1325         682 :         aw_pulse_set1(s, gb, block_idx, &fcb);
    1326         682 :         if (aw_pulse_set2(s, gb, block_idx, &fcb)) {
    1327             :             /* Conceal the block with silence and return.
    1328             :              * Skip the correct amount of bits to read the next
    1329             :              * block from the correct offset. */
    1330           0 :             int r_idx = pRNG(s->frame_cntr, block_idx, size);
    1331             : 
    1332           0 :             for (n = 0; n < size; n++)
    1333           0 :                 excitation[n] =
    1334           0 :                     wmavoice_std_codebook[r_idx + n] * s->silence_gain;
    1335           0 :             skip_bits(gb, 7 + 1);
    1336           0 :             return;
    1337             :         }
    1338             :     } else /* FCB_TYPE_EXC_PULSES */ {
    1339        9058 :         int offset_nbits = 5 - frame_desc->log_n_blocks;
    1340             : 
    1341        9058 :         fcb.no_repeat_mask = -1;
    1342             :         /* similar to ff_decode_10_pulses_35bits(), but with single pulses
    1343             :          * (instead of double) for a subset of pulses */
    1344       54348 :         for (n = 0; n < 5; n++) {
    1345             :             float sign;
    1346             :             int pos1, pos2;
    1347             : 
    1348       45290 :             sign           = get_bits1(gb) ? 1.0 : -1.0;
    1349       45290 :             pos1           = get_bits(gb, offset_nbits);
    1350       45290 :             fcb.x[fcb.n]   = n + 5 * pos1;
    1351       45290 :             fcb.y[fcb.n++] = sign;
    1352       45290 :             if (n < frame_desc->dbl_pulses) {
    1353       36270 :                 pos2           = get_bits(gb, offset_nbits);
    1354       36270 :                 fcb.x[fcb.n]   = n + 5 * pos2;
    1355       36270 :                 fcb.y[fcb.n++] = (pos1 < pos2) ? -sign : sign;
    1356             :             }
    1357             :         }
    1358             :     }
    1359        9740 :     ff_set_fixed_vector(pulses, &fcb, 1.0, size);
    1360             : 
    1361             :     /* Calculate gain for adaptive & fixed codebook signal.
    1362             :      * see ff_amr_set_fixed_gain(). */
    1363        9740 :     idx = get_bits(gb, 7);
    1364       19480 :     fcb_gain = expf(avpriv_scalarproduct_float_c(s->gain_pred_err,
    1365        9740 :                                                  gain_coeff, 6) -
    1366        9740 :                     5.2409161640 + wmavoice_gain_codebook_fcb[idx]);
    1367        9740 :     acb_gain = wmavoice_gain_codebook_acb[idx];
    1368        9740 :     pred_err = av_clipf(wmavoice_gain_codebook_fcb[idx],
    1369             :                         -2.9957322736 /* log(0.05) */,
    1370             :                          1.6094379124 /* log(5.0)  */);
    1371             : 
    1372        9740 :     gain_weight = 8 >> frame_desc->log_n_blocks;
    1373        9740 :     memmove(&s->gain_pred_err[gain_weight], s->gain_pred_err,
    1374        9740 :             sizeof(*s->gain_pred_err) * (6 - gain_weight));
    1375       30020 :     for (n = 0; n < gain_weight; n++)
    1376       20280 :         s->gain_pred_err[n] = pred_err;
    1377             : 
    1378             :     /* Calculation of adaptive codebook */
    1379        9740 :     if (frame_desc->acb_type == ACB_TYPE_ASYMMETRIC) {
    1380             :         int len;
    1381       19152 :         for (n = 0; n < size; n += len) {
    1382             :             int next_idx_sh16;
    1383       17876 :             int abs_idx    = block_idx * size + n;
    1384       35752 :             int pitch_sh16 = (s->last_pitch_val << 16) +
    1385       17876 :                              s->pitch_diff_sh16 * abs_idx;
    1386       17876 :             int pitch      = (pitch_sh16 + 0x6FFF) >> 16;
    1387       17876 :             int idx_sh16   = ((pitch << 16) - pitch_sh16) * 8 + 0x58000;
    1388       17876 :             idx            = idx_sh16 >> 16;
    1389       17876 :             if (s->pitch_diff_sh16) {
    1390       17442 :                 if (s->pitch_diff_sh16 > 0) {
    1391       10526 :                     next_idx_sh16 = (idx_sh16) &~ 0xFFFF;
    1392             :                 } else
    1393        6916 :                     next_idx_sh16 = (idx_sh16 + 0x10000) &~ 0xFFFF;
    1394       17442 :                 len = av_clip((idx_sh16 - next_idx_sh16) / s->pitch_diff_sh16 / 8,
    1395             :                               1, size - n);
    1396             :             } else
    1397         434 :                 len = size;
    1398             : 
    1399       17876 :             ff_acelp_interpolatef(&excitation[n], &excitation[n - pitch],
    1400             :                                   wmavoice_ipol1_coeffs, 17,
    1401             :                                   idx, 9, len);
    1402             :         }
    1403             :     } else /* ACB_TYPE_HAMMING */ {
    1404        8464 :         int block_pitch = block_pitch_sh2 >> 2;
    1405        8464 :         idx             = block_pitch_sh2 & 3;
    1406        8464 :         if (idx) {
    1407        3652 :             ff_acelp_interpolatef(excitation, &excitation[-block_pitch],
    1408             :                                   wmavoice_ipol2_coeffs, 4,
    1409             :                                   idx, 8, size);
    1410             :         } else
    1411        4812 :             av_memcpy_backptr((uint8_t *) excitation, sizeof(float) * block_pitch,
    1412             :                               sizeof(float) * size);
    1413             :     }
    1414             : 
    1415             :     /* Interpolate ACB/FCB and use as excitation signal */
    1416        9740 :     ff_weighted_vector_sumf(excitation, excitation, pulses,
    1417             :                             acb_gain, fcb_gain, size);
    1418             : }
    1419             : 
    1420             : /**
    1421             :  * Parse data in a single block.
    1422             :  *
    1423             :  * @param s WMA Voice decoding context private data
    1424             :  * @param gb bit I/O context
    1425             :  * @param block_idx index of the to-be-read block
    1426             :  * @param size amount of samples to be read in this block
    1427             :  * @param block_pitch_sh2 pitch for this block << 2
    1428             :  * @param lsps LSPs for (the end of) this frame
    1429             :  * @param prev_lsps LSPs for the last frame
    1430             :  * @param frame_desc frame type descriptor
    1431             :  * @param excitation target memory for the ACB+FCB interpolated signal
    1432             :  * @param synth target memory for the speech synthesis filter output
    1433             :  * @return 0 on success, <0 on error.
    1434             :  */
    1435       10783 : static void synth_block(WMAVoiceContext *s, GetBitContext *gb,
    1436             :                         int block_idx, int size,
    1437             :                         int block_pitch_sh2,
    1438             :                         const double *lsps, const double *prev_lsps,
    1439             :                         const struct frame_type_desc *frame_desc,
    1440             :                         float *excitation, float *synth)
    1441             : {
    1442             :     double i_lsps[MAX_LSPS];
    1443             :     float lpcs[MAX_LSPS];
    1444             :     float fac;
    1445             :     int n;
    1446             : 
    1447       10783 :     if (frame_desc->acb_type == ACB_TYPE_NONE)
    1448        1043 :         synth_block_hardcoded(s, gb, block_idx, size, frame_desc, excitation);
    1449             :     else
    1450        9740 :         synth_block_fcb_acb(s, gb, block_idx, size, block_pitch_sh2,
    1451             :                             frame_desc, excitation);
    1452             : 
    1453             :     /* convert interpolated LSPs to LPCs */
    1454       10783 :     fac = (block_idx + 0.5) / frame_desc->n_blocks;
    1455      151559 :     for (n = 0; n < s->lsps; n++) // LSF -> LSP
    1456      140776 :         i_lsps[n] = cos(prev_lsps[n] + fac * (lsps[n] - prev_lsps[n]));
    1457       10783 :     ff_acelp_lspd2lpc(i_lsps, lpcs, s->lsps >> 1);
    1458             : 
    1459             :     /* Speech synthesis */
    1460       10783 :     ff_celp_lp_synthesis_filterf(synth, lpcs, excitation, size, s->lsps);
    1461       10783 : }
    1462             : 
    1463             : /**
    1464             :  * Synthesize output samples for a single frame.
    1465             :  *
    1466             :  * @param ctx WMA Voice decoder context
    1467             :  * @param gb bit I/O context (s->gb or one for cross-packet superframes)
    1468             :  * @param frame_idx Frame number within superframe [0-2]
    1469             :  * @param samples pointer to output sample buffer, has space for at least 160
    1470             :  *                samples
    1471             :  * @param lsps LSP array
    1472             :  * @param prev_lsps array of previous frame's LSPs
    1473             :  * @param excitation target buffer for excitation signal
    1474             :  * @param synth target buffer for synthesized speech data
    1475             :  * @return 0 on success, <0 on error.
    1476             :  */
    1477        3306 : static int synth_frame(AVCodecContext *ctx, GetBitContext *gb, int frame_idx,
    1478             :                        float *samples,
    1479             :                        const double *lsps, const double *prev_lsps,
    1480             :                        float *excitation, float *synth)
    1481             : {
    1482        3306 :     WMAVoiceContext *s = ctx->priv_data;
    1483        3306 :     int n, n_blocks_x2, log_n_blocks_x2, av_uninit(cur_pitch_val);
    1484        3306 :     int pitch[MAX_BLOCKS], av_uninit(last_block_pitch);
    1485             : 
    1486             :     /* Parse frame type ("frame header"), see frame_descs */
    1487        3306 :     int bd_idx = s->vbm_tree[get_vlc2(gb, frame_type_vlc.table, 6, 3)], block_nsamples;
    1488             : 
    1489        3306 :     if (bd_idx < 0) {
    1490           0 :         av_log(ctx, AV_LOG_ERROR,
    1491             :                "Invalid frame type VLC code, skipping\n");
    1492           0 :         return AVERROR_INVALIDDATA;
    1493             :     }
    1494             : 
    1495        3306 :     block_nsamples = MAX_FRAMESIZE / frame_descs[bd_idx].n_blocks;
    1496             : 
    1497             :     /* Pitch calculation for ACB_TYPE_ASYMMETRIC ("pitch-per-frame") */
    1498        3306 :     if (frame_descs[bd_idx].acb_type == ACB_TYPE_ASYMMETRIC) {
    1499             :         /* Pitch is provided per frame, which is interpreted as the pitch of
    1500             :          * the last sample of the last block of this frame. We can interpolate
    1501             :          * the pitch of other blocks (and even pitch-per-sample) by gradually
    1502             :          * incrementing/decrementing prev_frame_pitch to cur_pitch_val. */
    1503         560 :         n_blocks_x2      = frame_descs[bd_idx].n_blocks << 1;
    1504         560 :         log_n_blocks_x2  = frame_descs[bd_idx].log_n_blocks + 1;
    1505         560 :         cur_pitch_val    = s->min_pitch_val + get_bits(gb, s->pitch_nbits);
    1506         560 :         cur_pitch_val    = FFMIN(cur_pitch_val, s->max_pitch_val - 1);
    1507        1084 :         if (s->last_acb_type == ACB_TYPE_NONE ||
    1508         524 :             20 * abs(cur_pitch_val - s->last_pitch_val) >
    1509         524 :                 (cur_pitch_val + s->last_pitch_val))
    1510         138 :             s->last_pitch_val = cur_pitch_val;
    1511             : 
    1512             :         /* pitch per block */
    1513        1836 :         for (n = 0; n < frame_descs[bd_idx].n_blocks; n++) {
    1514        1276 :             int fac = n * 2 + 1;
    1515             : 
    1516        3828 :             pitch[n] = (MUL16(fac,                 cur_pitch_val) +
    1517        2552 :                         MUL16((n_blocks_x2 - fac), s->last_pitch_val) +
    1518        2552 :                         frame_descs[bd_idx].n_blocks) >> log_n_blocks_x2;
    1519             :         }
    1520             : 
    1521             :         /* "pitch-diff-per-sample" for calculation of pitch per sample */
    1522         560 :         s->pitch_diff_sh16 =
    1523         560 :             ((cur_pitch_val - s->last_pitch_val) << 16) / MAX_FRAMESIZE;
    1524             :     }
    1525             : 
    1526             :     /* Global gain (if silence) and pitch-adaptive window coordinates */
    1527        3306 :     switch (frame_descs[bd_idx].fcb_type) {
    1528         499 :     case FCB_TYPE_SILENCE:
    1529         499 :         s->silence_gain = wmavoice_gain_silence[get_bits(gb, 8)];
    1530         499 :         break;
    1531         341 :     case FCB_TYPE_AW_PULSES:
    1532         341 :         aw_parse_coords(s, gb, pitch);
    1533         341 :         break;
    1534             :     }
    1535             : 
    1536       14089 :     for (n = 0; n < frame_descs[bd_idx].n_blocks; n++) {
    1537             :         int bl_pitch_sh2;
    1538             : 
    1539             :         /* Pitch calculation for ACB_TYPE_HAMMING ("pitch-per-block") */
    1540       10783 :         switch (frame_descs[bd_idx].acb_type) {
    1541        8464 :         case ACB_TYPE_HAMMING: {
    1542             :             /* Pitch is given per block. Per-block pitches are encoded as an
    1543             :              * absolute value for the first block, and then delta values
    1544             :              * relative to this value) for all subsequent blocks. The scale of
    1545             :              * this pitch value is semi-logarithmic compared to its use in the
    1546             :              * decoder, so we convert it to normal scale also. */
    1547             :             int block_pitch,
    1548        8464 :                 t1 = (s->block_conv_table[1] - s->block_conv_table[0]) << 2,
    1549        8464 :                 t2 = (s->block_conv_table[2] - s->block_conv_table[1]) << 1,
    1550        8464 :                 t3 =  s->block_conv_table[3] - s->block_conv_table[2] + 1;
    1551             : 
    1552        8464 :             if (n == 0) {
    1553        1975 :                 block_pitch = get_bits(gb, s->block_pitch_nbits);
    1554             :             } else
    1555       12978 :                 block_pitch = last_block_pitch - s->block_delta_pitch_hrange +
    1556        6489 :                                  get_bits(gb, s->block_delta_pitch_nbits);
    1557             :             /* Convert last_ so that any next delta is within _range */
    1558        8464 :             last_block_pitch = av_clip(block_pitch,
    1559             :                                        s->block_delta_pitch_hrange,
    1560        8464 :                                        s->block_pitch_range -
    1561        8464 :                                            s->block_delta_pitch_hrange);
    1562             : 
    1563             :             /* Convert semi-log-style scale back to normal scale */
    1564        8464 :             if (block_pitch < t1) {
    1565        1491 :                 bl_pitch_sh2 = (s->block_conv_table[0] << 2) + block_pitch;
    1566             :             } else {
    1567        6973 :                 block_pitch -= t1;
    1568        6973 :                 if (block_pitch < t2) {
    1569        5712 :                     bl_pitch_sh2 =
    1570        5712 :                         (s->block_conv_table[1] << 2) + (block_pitch << 1);
    1571             :                 } else {
    1572        1261 :                     block_pitch -= t2;
    1573        1261 :                     if (block_pitch < t3) {
    1574        1261 :                         bl_pitch_sh2 =
    1575        1261 :                             (s->block_conv_table[2] + block_pitch) << 2;
    1576             :                     } else
    1577           0 :                         bl_pitch_sh2 = s->block_conv_table[3] << 2;
    1578             :                 }
    1579             :             }
    1580        8464 :             pitch[n] = bl_pitch_sh2 >> 2;
    1581        8464 :             break;
    1582             :         }
    1583             : 
    1584        1276 :         case ACB_TYPE_ASYMMETRIC: {
    1585        1276 :             bl_pitch_sh2 = pitch[n] << 2;
    1586        1276 :             break;
    1587             :         }
    1588             : 
    1589        1043 :         default: // ACB_TYPE_NONE has no pitch
    1590        1043 :             bl_pitch_sh2 = 0;
    1591        1043 :             break;
    1592             :         }
    1593             : 
    1594       21566 :         synth_block(s, gb, n, block_nsamples, bl_pitch_sh2,
    1595             :                     lsps, prev_lsps, &frame_descs[bd_idx],
    1596       10783 :                     &excitation[n * block_nsamples],
    1597       10783 :                     &synth[n * block_nsamples]);
    1598             :     }
    1599             : 
    1600             :     /* Averaging projection filter, if applicable. Else, just copy samples
    1601             :      * from synthesis buffer */
    1602        3306 :     if (s->do_apf) {
    1603             :         double i_lsps[MAX_LSPS];
    1604             :         float lpcs[MAX_LSPS];
    1605             : 
    1606       46266 :         for (n = 0; n < s->lsps; n++) // LSF -> LSP
    1607       42960 :             i_lsps[n] = cos(0.5 * (prev_lsps[n] + lsps[n]));
    1608        3306 :         ff_acelp_lspd2lpc(i_lsps, lpcs, s->lsps >> 1);
    1609        9918 :         postfilter(s, synth, samples, 80, lpcs,
    1610        3306 :                    &s->zero_exc_pf[s->history_nsamples + MAX_FRAMESIZE * frame_idx],
    1611        3306 :                    frame_descs[bd_idx].fcb_type, pitch[0]);
    1612             : 
    1613       46266 :         for (n = 0; n < s->lsps; n++) // LSF -> LSP
    1614       42960 :             i_lsps[n] = cos(lsps[n]);
    1615        3306 :         ff_acelp_lspd2lpc(i_lsps, lpcs, s->lsps >> 1);
    1616        9918 :         postfilter(s, &synth[80], &samples[80], 80, lpcs,
    1617        3306 :                    &s->zero_exc_pf[s->history_nsamples + MAX_FRAMESIZE * frame_idx + 80],
    1618        3306 :                    frame_descs[bd_idx].fcb_type, pitch[0]);
    1619             :     } else
    1620           0 :         memcpy(samples, synth, 160 * sizeof(synth[0]));
    1621             : 
    1622             :     /* Cache values for next frame */
    1623        3306 :     s->frame_cntr++;
    1624        3306 :     if (s->frame_cntr >= 0xFFFF) s->frame_cntr -= 0xFFFF; // i.e. modulo (%)
    1625        3306 :     s->last_acb_type = frame_descs[bd_idx].acb_type;
    1626        3306 :     switch (frame_descs[bd_idx].acb_type) {
    1627         771 :     case ACB_TYPE_NONE:
    1628         771 :         s->last_pitch_val = 0;
    1629         771 :         break;
    1630         560 :     case ACB_TYPE_ASYMMETRIC:
    1631         560 :         s->last_pitch_val = cur_pitch_val;
    1632         560 :         break;
    1633        1975 :     case ACB_TYPE_HAMMING:
    1634        1975 :         s->last_pitch_val = pitch[frame_descs[bd_idx].n_blocks - 1];
    1635        1975 :         break;
    1636             :     }
    1637             : 
    1638        3306 :     return 0;
    1639             : }
    1640             : 
    1641             : /**
    1642             :  * Ensure minimum value for first item, maximum value for last value,
    1643             :  * proper spacing between each value and proper ordering.
    1644             :  *
    1645             :  * @param lsps array of LSPs
    1646             :  * @param num size of LSP array
    1647             :  *
    1648             :  * @note basically a double version of #ff_acelp_reorder_lsf(), might be
    1649             :  *       useful to put in a generic location later on. Parts are also
    1650             :  *       present in #ff_set_min_dist_lsf() + #ff_sort_nearly_sorted_floats(),
    1651             :  *       which is in float.
    1652             :  */
    1653        3306 : static void stabilize_lsps(double *lsps, int num)
    1654             : {
    1655             :     int n, m, l;
    1656             : 
    1657             :     /* set minimum value for first, maximum value for last and minimum
    1658             :      * spacing between LSF values.
    1659             :      * Very similar to ff_set_min_dist_lsf(), but in double. */
    1660        3306 :     lsps[0]       = FFMAX(lsps[0],       0.0015 * M_PI);
    1661       42960 :     for (n = 1; n < num; n++)
    1662       39654 :         lsps[n]   = FFMAX(lsps[n],       lsps[n - 1] + 0.0125 * M_PI);
    1663        3306 :     lsps[num - 1] = FFMIN(lsps[num - 1], 0.9985 * M_PI);
    1664             : 
    1665             :     /* reorder (looks like one-time / non-recursed bubblesort).
    1666             :      * Very similar to ff_sort_nearly_sorted_floats(), but in double. */
    1667       42960 :     for (n = 1; n < num; n++) {
    1668       39654 :         if (lsps[n] < lsps[n - 1]) {
    1669           0 :             for (m = 1; m < num; m++) {
    1670           0 :                 double tmp = lsps[m];
    1671           0 :                 for (l = m - 1; l >= 0; l--) {
    1672           0 :                     if (lsps[l] <= tmp) break;
    1673           0 :                     lsps[l + 1] = lsps[l];
    1674             :                 }
    1675           0 :                 lsps[l + 1] = tmp;
    1676             :             }
    1677           0 :             break;
    1678             :         }
    1679             :     }
    1680        3306 : }
    1681             : 
    1682             : /**
    1683             :  * Synthesize output samples for a single superframe. If we have any data
    1684             :  * cached in s->sframe_cache, that will be used instead of whatever is loaded
    1685             :  * in s->gb.
    1686             :  *
    1687             :  * WMA Voice superframes contain 3 frames, each containing 160 audio samples,
    1688             :  * to give a total of 480 samples per frame. See #synth_frame() for frame
    1689             :  * parsing. In addition to 3 frames, superframes can also contain the LSPs
    1690             :  * (if these are globally specified for all frames (residually); they can
    1691             :  * also be specified individually per-frame. See the s->has_residual_lsps
    1692             :  * option), and can specify the number of samples encoded in this superframe
    1693             :  * (if less than 480), usually used to prevent blanks at track boundaries.
    1694             :  *
    1695             :  * @param ctx WMA Voice decoder context
    1696             :  * @return 0 on success, <0 on error or 1 if there was not enough data to
    1697             :  *         fully parse the superframe
    1698             :  */
    1699        1102 : static int synth_superframe(AVCodecContext *ctx, AVFrame *frame,
    1700             :                             int *got_frame_ptr)
    1701             : {
    1702        1102 :     WMAVoiceContext *s = ctx->priv_data;
    1703        1102 :     GetBitContext *gb = &s->gb, s_gb;
    1704        1102 :     int n, res, n_samples = MAX_SFRAMESIZE;
    1705             :     double lsps[MAX_FRAMES][MAX_LSPS];
    1706        2204 :     const double *mean_lsf = s->lsps == 16 ?
    1707        1102 :         wmavoice_mean_lsf16[s->lsp_def_mode] : wmavoice_mean_lsf10[s->lsp_def_mode];
    1708             :     float excitation[MAX_SIGNAL_HISTORY + MAX_SFRAMESIZE + 12];
    1709             :     float synth[MAX_LSPS + MAX_SFRAMESIZE];
    1710             :     float *samples;
    1711             : 
    1712        1102 :     memcpy(synth,      s->synth_history,
    1713        1102 :            s->lsps             * sizeof(*synth));
    1714        1102 :     memcpy(excitation, s->excitation_history,
    1715        1102 :            s->history_nsamples * sizeof(*excitation));
    1716             : 
    1717        1102 :     if (s->sframe_cache_size > 0) {
    1718         185 :         gb = &s_gb;
    1719         185 :         init_get_bits(gb, s->sframe_cache, s->sframe_cache_size);
    1720         185 :         s->sframe_cache_size = 0;
    1721             :     }
    1722             : 
    1723             :     /* First bit is speech/music bit, it differentiates between WMAVoice
    1724             :      * speech samples (the actual codec) and WMAVoice music samples, which
    1725             :      * are really WMAPro-in-WMAVoice-superframes. I've never seen those in
    1726             :      * the wild yet. */
    1727        1102 :     if (!get_bits1(gb)) {
    1728           0 :         avpriv_request_sample(ctx, "WMAPro-in-WMAVoice");
    1729           0 :         return AVERROR_PATCHWELCOME;
    1730             :     }
    1731             : 
    1732             :     /* (optional) nr. of samples in superframe; always <= 480 and >= 0 */
    1733        1102 :     if (get_bits1(gb)) {
    1734           3 :         if ((n_samples = get_bits(gb, 12)) > MAX_SFRAMESIZE) {
    1735           0 :             av_log(ctx, AV_LOG_ERROR,
    1736             :                    "Superframe encodes > %d samples (%d), not allowed\n",
    1737             :                    MAX_SFRAMESIZE, n_samples);
    1738           0 :             return AVERROR_INVALIDDATA;
    1739             :         }
    1740             :     }
    1741             : 
    1742             :     /* Parse LSPs, if global for the superframe (can also be per-frame). */
    1743        1102 :     if (s->has_residual_lsps) {
    1744             :         double prev_lsps[MAX_LSPS], a1[MAX_LSPS * 2], a2[MAX_LSPS * 2];
    1745             : 
    1746       15422 :         for (n = 0; n < s->lsps; n++)
    1747       14320 :             prev_lsps[n] = s->prev_lsps[n] - mean_lsf[n];
    1748             : 
    1749        1102 :         if (s->lsps == 10) {
    1750         552 :             dequant_lsp10r(gb, lsps[2], prev_lsps, a1, a2, s->lsp_q_mode);
    1751             :         } else /* s->lsps == 16 */
    1752         550 :             dequant_lsp16r(gb, lsps[2], prev_lsps, a1, a2, s->lsp_q_mode);
    1753             : 
    1754       15422 :         for (n = 0; n < s->lsps; n++) {
    1755       14320 :             lsps[0][n]  = mean_lsf[n] + (a1[n]           - a2[n * 2]);
    1756       14320 :             lsps[1][n]  = mean_lsf[n] + (a1[s->lsps + n] - a2[n * 2 + 1]);
    1757       14320 :             lsps[2][n] += mean_lsf[n];
    1758             :         }
    1759        4408 :         for (n = 0; n < 3; n++)
    1760        3306 :             stabilize_lsps(lsps[n], s->lsps);
    1761             :     }
    1762             : 
    1763             :     /* synth_superframe can run multiple times per packet
    1764             :      * free potential previous frame */
    1765        1102 :     av_frame_unref(frame);
    1766             : 
    1767             :     /* get output buffer */
    1768        1102 :     frame->nb_samples = MAX_SFRAMESIZE;
    1769        1102 :     if ((res = ff_get_buffer(ctx, frame, 0)) < 0)
    1770           0 :         return res;
    1771        1102 :     frame->nb_samples = n_samples;
    1772        1102 :     samples = (float *)frame->data[0];
    1773             : 
    1774             :     /* Parse frames, optionally preceded by per-frame (independent) LSPs. */
    1775        4408 :     for (n = 0; n < 3; n++) {
    1776        3306 :         if (!s->has_residual_lsps) {
    1777             :             int m;
    1778             : 
    1779           0 :             if (s->lsps == 10) {
    1780           0 :                 dequant_lsp10i(gb, lsps[n]);
    1781             :             } else /* s->lsps == 16 */
    1782           0 :                 dequant_lsp16i(gb, lsps[n]);
    1783             : 
    1784           0 :             for (m = 0; m < s->lsps; m++)
    1785           0 :                 lsps[n][m] += mean_lsf[m];
    1786           0 :             stabilize_lsps(lsps[n], s->lsps);
    1787             :         }
    1788             : 
    1789       12122 :         if ((res = synth_frame(ctx, gb, n,
    1790             :                                &samples[n * MAX_FRAMESIZE],
    1791        5510 :                                lsps[n], n == 0 ? s->prev_lsps : lsps[n - 1],
    1792        3306 :                                &excitation[s->history_nsamples + n * MAX_FRAMESIZE],
    1793        3306 :                                &synth[s->lsps + n * MAX_FRAMESIZE]))) {
    1794           0 :             *got_frame_ptr = 0;
    1795           0 :             return res;
    1796             :         }
    1797             :     }
    1798             : 
    1799             :     /* Statistics? FIXME - we don't check for length, a slight overrun
    1800             :      * will be caught by internal buffer padding, and anything else
    1801             :      * will be skipped, not read. */
    1802        1102 :     if (get_bits1(gb)) {
    1803           0 :         res = get_bits(gb, 4);
    1804           0 :         skip_bits(gb, 10 * (res + 1));
    1805             :     }
    1806             : 
    1807        1102 :     if (get_bits_left(gb) < 0) {
    1808           0 :         wmavoice_flush(ctx);
    1809           0 :         return AVERROR_INVALIDDATA;
    1810             :     }
    1811             : 
    1812        1102 :     *got_frame_ptr = 1;
    1813             : 
    1814             :     /* Update history */
    1815        1102 :     memcpy(s->prev_lsps,           lsps[2],
    1816        1102 :            s->lsps             * sizeof(*s->prev_lsps));
    1817        1102 :     memcpy(s->synth_history,      &synth[MAX_SFRAMESIZE],
    1818        1102 :            s->lsps             * sizeof(*synth));
    1819        1102 :     memcpy(s->excitation_history, &excitation[MAX_SFRAMESIZE],
    1820        1102 :            s->history_nsamples * sizeof(*excitation));
    1821        1102 :     if (s->do_apf)
    1822        1102 :         memmove(s->zero_exc_pf,       &s->zero_exc_pf[MAX_SFRAMESIZE],
    1823        1102 :                 s->history_nsamples * sizeof(*s->zero_exc_pf));
    1824             : 
    1825        1102 :     return 0;
    1826             : }
    1827             : 
    1828             : /**
    1829             :  * Parse the packet header at the start of each packet (input data to this
    1830             :  * decoder).
    1831             :  *
    1832             :  * @param s WMA Voice decoding context private data
    1833             :  * @return <0 on error, nb_superframes on success.
    1834             :  */
    1835         186 : static int parse_packet_header(WMAVoiceContext *s)
    1836             : {
    1837         186 :     GetBitContext *gb = &s->gb;
    1838         186 :     unsigned int res, n_superframes = 0;
    1839             : 
    1840         186 :     skip_bits(gb, 4);          // packet sequence number
    1841         186 :     s->has_residual_lsps = get_bits1(gb);
    1842             :     do {
    1843         186 :         res = get_bits(gb, 6); // number of superframes per packet
    1844             :                                // (minus first one if there is spillover)
    1845         186 :         n_superframes += res;
    1846         186 :     } while (res == 0x3F);
    1847         186 :     s->spillover_nbits   = get_bits(gb, s->spillover_bitsize);
    1848             : 
    1849         186 :     return get_bits_left(gb) >= 0 ? n_superframes : AVERROR_INVALIDDATA;
    1850             : }
    1851             : 
    1852             : /**
    1853             :  * Copy (unaligned) bits from gb/data/size to pb.
    1854             :  *
    1855             :  * @param pb target buffer to copy bits into
    1856             :  * @param data source buffer to copy bits from
    1857             :  * @param size size of the source data, in bytes
    1858             :  * @param gb bit I/O context specifying the current position in the source.
    1859             :  *           data. This function might use this to align the bit position to
    1860             :  *           a whole-byte boundary before calling #avpriv_copy_bits() on aligned
    1861             :  *           source data
    1862             :  * @param nbits the amount of bits to copy from source to target
    1863             :  *
    1864             :  * @note after calling this function, the current position in the input bit
    1865             :  *       I/O context is undefined.
    1866             :  */
    1867         370 : static void copy_bits(PutBitContext *pb,
    1868             :                       const uint8_t *data, int size,
    1869             :                       GetBitContext *gb, int nbits)
    1870             : {
    1871             :     int rmn_bytes, rmn_bits;
    1872             : 
    1873         370 :     rmn_bits = rmn_bytes = get_bits_left(gb);
    1874         370 :     if (rmn_bits < nbits)
    1875           0 :         return;
    1876         370 :     if (nbits > pb->size_in_bits - put_bits_count(pb))
    1877           0 :         return;
    1878         370 :     rmn_bits &= 7; rmn_bytes >>= 3;
    1879         370 :     if ((rmn_bits = FFMIN(rmn_bits, nbits)) > 0)
    1880         290 :         put_bits(pb, rmn_bits, get_bits(gb, rmn_bits));
    1881         370 :     avpriv_copy_bits(pb, data + size - rmn_bytes,
    1882         370 :                  FFMIN(nbits - rmn_bits, rmn_bytes << 3));
    1883             : }
    1884             : 
    1885             : /**
    1886             :  * Packet decoding: a packet is anything that the (ASF) demuxer contains,
    1887             :  * and we expect that the demuxer / application provides it to us as such
    1888             :  * (else you'll probably get garbage as output). Every packet has a size of
    1889             :  * ctx->block_align bytes, starts with a packet header (see
    1890             :  * #parse_packet_header()), and then a series of superframes. Superframe
    1891             :  * boundaries may exceed packets, i.e. superframes can split data over
    1892             :  * multiple (two) packets.
    1893             :  *
    1894             :  * For more information about frames, see #synth_superframe().
    1895             :  */
    1896        1291 : static int wmavoice_decode_packet(AVCodecContext *ctx, void *data,
    1897             :                                   int *got_frame_ptr, AVPacket *avpkt)
    1898             : {
    1899        1291 :     WMAVoiceContext *s = ctx->priv_data;
    1900        1291 :     GetBitContext *gb = &s->gb;
    1901             :     int size, res, pos;
    1902             : 
    1903             :     /* Packets are sometimes a multiple of ctx->block_align, with a packet
    1904             :      * header at each ctx->block_align bytes. However, FFmpeg's ASF demuxer
    1905             :      * feeds us ASF packets, which may concatenate multiple "codec" packets
    1906             :      * in a single "muxer" packet, so we artificially emulate that by
    1907             :      * capping the packet size at ctx->block_align. */
    1908        1291 :     for (size = avpkt->size; size > ctx->block_align; size -= ctx->block_align);
    1909        1291 :     init_get_bits(&s->gb, avpkt->data, size << 3);
    1910             : 
    1911             :     /* size == ctx->block_align is used to indicate whether we are dealing with
    1912             :      * a new packet or a packet of which we already read the packet header
    1913             :      * previously. */
    1914        1291 :     if (!(size % ctx->block_align)) { // new packet header
    1915         191 :         if (!size) {
    1916           5 :             s->spillover_nbits = 0;
    1917           5 :             s->nb_superframes = 0;
    1918             :         } else {
    1919         186 :             if ((res = parse_packet_header(s)) < 0)
    1920           0 :                 return res;
    1921         186 :             s->nb_superframes = res;
    1922             :         }
    1923             : 
    1924             :         /* If the packet header specifies a s->spillover_nbits, then we want
    1925             :          * to push out all data of the previous packet (+ spillover) before
    1926             :          * continuing to parse new superframes in the current packet. */
    1927         191 :         if (s->sframe_cache_size > 0) {
    1928         185 :             int cnt = get_bits_count(gb);
    1929         185 :             if (cnt + s->spillover_nbits > avpkt->size * 8) {
    1930           0 :                 s->spillover_nbits = avpkt->size * 8 - cnt;
    1931             :             }
    1932         185 :             copy_bits(&s->pb, avpkt->data, size, gb, s->spillover_nbits);
    1933         185 :             flush_put_bits(&s->pb);
    1934         185 :             s->sframe_cache_size += s->spillover_nbits;
    1935         370 :             if ((res = synth_superframe(ctx, data, got_frame_ptr)) == 0 &&
    1936         185 :                 *got_frame_ptr) {
    1937         185 :                 cnt += s->spillover_nbits;
    1938         185 :                 s->skip_bits_next = cnt & 7;
    1939         185 :                 res = cnt >> 3;
    1940         185 :                 return res;
    1941             :             } else
    1942           0 :                 skip_bits_long (gb, s->spillover_nbits - cnt +
    1943           0 :                                 get_bits_count(gb)); // resync
    1944           6 :         } else if (s->spillover_nbits) {
    1945           0 :             skip_bits_long(gb, s->spillover_nbits);  // resync
    1946             :         }
    1947        1100 :     } else if (s->skip_bits_next)
    1948         971 :         skip_bits(gb, s->skip_bits_next);
    1949             : 
    1950             :     /* Try parsing superframes in current packet */
    1951        1106 :     s->sframe_cache_size = 0;
    1952        1106 :     s->skip_bits_next = 0;
    1953        1106 :     pos = get_bits_left(gb);
    1954        1106 :     if (s->nb_superframes-- == 0) {
    1955           4 :         *got_frame_ptr = 0;
    1956           4 :         return size;
    1957        1102 :     } else if (s->nb_superframes > 0) {
    1958         917 :         if ((res = synth_superframe(ctx, data, got_frame_ptr)) < 0) {
    1959           0 :             return res;
    1960         917 :         } else if (*got_frame_ptr) {
    1961         917 :             int cnt = get_bits_count(gb);
    1962         917 :             s->skip_bits_next = cnt & 7;
    1963         917 :             res = cnt >> 3;
    1964         917 :             return res;
    1965             :         }
    1966         185 :     } else if ((s->sframe_cache_size = pos) > 0) {
    1967             :         /* ... cache it for spillover in next packet */
    1968         185 :         init_put_bits(&s->pb, s->sframe_cache, SFRAME_CACHE_MAXSIZE);
    1969         185 :         copy_bits(&s->pb, avpkt->data, size, gb, s->sframe_cache_size);
    1970             :         // FIXME bad - just copy bytes as whole and add use the
    1971             :         // skip_bits_next field
    1972             :     }
    1973             : 
    1974         185 :     return size;
    1975             : }
    1976             : 
    1977           8 : static av_cold int wmavoice_decode_end(AVCodecContext *ctx)
    1978             : {
    1979           8 :     WMAVoiceContext *s = ctx->priv_data;
    1980             : 
    1981           8 :     if (s->do_apf) {
    1982           8 :         ff_rdft_end(&s->rdft);
    1983           8 :         ff_rdft_end(&s->irdft);
    1984           8 :         ff_dct_end(&s->dct);
    1985           8 :         ff_dct_end(&s->dst);
    1986             :     }
    1987             : 
    1988           8 :     return 0;
    1989             : }
    1990             : 
    1991             : AVCodec ff_wmavoice_decoder = {
    1992             :     .name             = "wmavoice",
    1993             :     .long_name        = NULL_IF_CONFIG_SMALL("Windows Media Audio Voice"),
    1994             :     .type             = AVMEDIA_TYPE_AUDIO,
    1995             :     .id               = AV_CODEC_ID_WMAVOICE,
    1996             :     .priv_data_size   = sizeof(WMAVoiceContext),
    1997             :     .init             = wmavoice_decode_init,
    1998             :     .close            = wmavoice_decode_end,
    1999             :     .decode           = wmavoice_decode_packet,
    2000             :     .capabilities     = AV_CODEC_CAP_SUBFRAMES | AV_CODEC_CAP_DR1 | AV_CODEC_CAP_DELAY,
    2001             :     .flush            = wmavoice_flush,
    2002             : };

Generated by: LCOV version 1.13