psymodel.h 5.5 KB
Newer Older
1 2 3 4
/*
 * audio encoder psychoacoustic model
 * Copyright (C) 2008 Konstantin Shishkov
 *
5
 * This file is part of Libav.
6
 *
7
 * Libav is free software; you can redistribute it and/or
8 9 10 11
 * modify it under the terms of the GNU Lesser General Public
 * License as published by the Free Software Foundation; either
 * version 2.1 of the License, or (at your option) any later version.
 *
12
 * Libav is distributed in the hope that it will be useful,
13 14 15 16 17
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
 * Lesser General Public License for more details.
 *
 * You should have received a copy of the GNU Lesser General Public
18
 * License along with Libav; if not, write to the Free Software
19 20 21 22 23 24 25 26 27 28
 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
 */

#ifndef AVCODEC_PSYMODEL_H
#define AVCODEC_PSYMODEL_H

#include "avcodec.h"

/** maximum possible number of bands */
#define PSY_MAX_BANDS 128
29 30
/** maximum number of channels */
#define PSY_MAX_CHANS 20
31 32 33 34

/**
 * single band psychoacoustic information
 */
35
typedef struct FFPsyBand {
36 37 38 39 40
    int   bits;
    float energy;
    float threshold;
    float distortion;
    float perceptual_weight;
41
} FFPsyBand;
42 43 44 45

/**
 * windowing related information
 */
46
typedef struct FFPsyWindowInfo {
47 48 49 50 51
    int window_type[3];               ///< window type (short/long/transitional, etc.) - current, previous and next
    int window_shape;                 ///< window shape (sine/KBD/whatever)
    int num_windows;                  ///< number of windows in a frame
    int grouping[8];                  ///< window grouping (for e.g. AAC)
    int *window_sizes;                ///< sequence of window sizes inside one frame (for eg. WMA)
52
} FFPsyWindowInfo;
53 54 55 56

/**
 * context used by psychoacoustic model
 */
57
typedef struct FFPsyContext {
58 59 60 61 62 63 64 65 66
    AVCodecContext *avctx;            ///< encoder context
    const struct FFPsyModel *model;   ///< encoder-specific model functions

    FFPsyBand *psy_bands;             ///< frame bands information

    uint8_t **bands;                  ///< scalefactor band sizes for possible frame sizes
    int     *num_bands;               ///< number of scalefactor bands for possible frame sizes
    int num_lens;                     ///< number of scalefactor band sets

67 68 69 70 71 72 73
    float pe[PSY_MAX_CHANS];          ///< total PE for each channel in the frame

    struct {
        int size;                     ///< size of the bitresevoir in bits
        int bits;                     ///< number of bits used in the bitresevoir
    } bitres;

74
    void* model_priv_data;            ///< psychoacoustic model implementation private data
75
} FFPsyContext;
76 77 78 79 80 81 82

/**
 * codec-specific psychoacoustic model implementation
 */
typedef struct FFPsyModel {
    const char *name;
    int  (*init)   (FFPsyContext *apc);
83 84 85 86 87 88 89 90 91 92 93 94

    /**
     * Suggest window sequence for channel.
     *
     * @param ctx       model context
     * @param audio     samples for the current frame
     * @param la        lookahead samples (NULL when unavailable)
     * @param channel   number of channel element to analyze
     * @param prev_type previous window type
     *
     * @return suggested window information in a structure
     */
95
    FFPsyWindowInfo (*window)(FFPsyContext *ctx, const int16_t *audio, const int16_t *la, int channel, int prev_type);
96 97 98 99 100 101 102 103 104

    /**
     * Perform psychoacoustic analysis and set band info (threshold, energy).
     *
     * @param ctx     model context
     * @param channel audio channel number
     * @param coeffs  pointer to the transformed coefficients
     * @param wi      window information
     */
105
    void (*analyze)(FFPsyContext *ctx, int channel, const float *coeffs, const FFPsyWindowInfo *wi);
106

107
    void (*end)    (FFPsyContext *apc);
108
} FFPsyModel;
109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162

/**
 * Initialize psychoacoustic model.
 *
 * @param ctx        model context
 * @param avctx      codec context
 * @param num_lens   number of possible frame lengths
 * @param bands      scalefactor band lengths for all frame lengths
 * @param num_bands  number of scalefactor bands for all frame lengths
 *
 * @return zero if successful, a negative value if not
 */
av_cold int ff_psy_init(FFPsyContext *ctx, AVCodecContext *avctx,
                        int num_lens,
                        const uint8_t **bands, const int* num_bands);

/**
 * Cleanup model context at the end.
 *
 * @param ctx model context
 */
av_cold void ff_psy_end(FFPsyContext *ctx);


/**************************************************************************
 *                       Audio preprocessing stuff.                       *
 *       This should be moved into some audio filter eventually.          *
 **************************************************************************/
struct FFPsyPreprocessContext;

/**
 * psychoacoustic model audio preprocessing initialization
 */
av_cold struct FFPsyPreprocessContext* ff_psy_preprocess_init(AVCodecContext *avctx);

/**
 * Preprocess several channel in audio frame in order to compress it better.
 *
 * @param ctx      preprocessing context
 * @param audio    samples to preprocess
 * @param dest     place to put filtered samples
 * @param tag      channel number
 * @param channels number of channel to preprocess (some additional work may be done on stereo pair)
 */
void ff_psy_preprocess(struct FFPsyPreprocessContext *ctx,
                       const int16_t *audio, int16_t *dest,
                       int tag, int channels);

/**
 * Cleanup audio preprocessing module.
 */
av_cold void ff_psy_preprocess_end(struct FFPsyPreprocessContext *ctx);

#endif /* AVCODEC_PSYMODEL_H */