psymodel.h 6.48 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 46 47 48 49 50 51 52 53 54 55 56 57 58 59
/**
 * single channel psychoacoustic information
 */
typedef struct FFPsyChannel {
    FFPsyBand psy_bands[PSY_MAX_BANDS]; ///< channel bands information
    float     entropy;                  ///< total PE for this channel
} FFPsyChannel;

/**
 * psychoacoustic information for an arbitrary group of channels
 */
typedef struct FFPsyChannelGroup {
    FFPsyChannel *ch[PSY_MAX_CHANS];  ///< pointers to the individual channels in the group
    uint8_t num_ch;                   ///< number of channels in this group
    uint8_t coupling[PSY_MAX_BANDS];  ///< allow coupling for this band in the group
} FFPsyChannelGroup;

60 61 62
/**
 * windowing related information
 */
63
typedef struct FFPsyWindowInfo {
64 65 66 67 68
    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)
69
} FFPsyWindowInfo;
70 71 72 73

/**
 * context used by psychoacoustic model
 */
74
typedef struct FFPsyContext {
75 76 77
    AVCodecContext *avctx;            ///< encoder context
    const struct FFPsyModel *model;   ///< encoder-specific model functions

78 79 80
    FFPsyChannel      *ch;            ///< single channel information
    FFPsyChannelGroup *group;         ///< channel group information
    int num_groups;                   ///< number of channel groups
81 82 83 84 85

    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

86 87 88 89 90
    struct {
        int size;                     ///< size of the bitresevoir in bits
        int bits;                     ///< number of bits used in the bitresevoir
    } bitres;

91
    void* model_priv_data;            ///< psychoacoustic model implementation private data
92
} FFPsyContext;
93 94 95 96 97 98 99

/**
 * codec-specific psychoacoustic model implementation
 */
typedef struct FFPsyModel {
    const char *name;
    int  (*init)   (FFPsyContext *apc);
100 101 102 103 104 105 106 107 108 109 110 111

    /**
     * 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
     */
112
    FFPsyWindowInfo (*window)(FFPsyContext *ctx, const float *audio, const float *la, int channel, int prev_type);
113

114 115 116 117 118 119 120 121
    /**
     * Perform psychoacoustic analysis and set band info (threshold, energy) for a group of channels.
     *
     * @param ctx      model context
     * @param channel  channel number of the first channel in the group to perform analysis on
     * @param coeffs   array of pointers to the transformed coefficients
     * @param wi       window information for the channels in the group
     */
122
    void (*analyze)(FFPsyContext *ctx, int channel, const float **coeffs, const FFPsyWindowInfo *wi);
123

124
    void (*end)    (FFPsyContext *apc);
125
} FFPsyModel;
126 127 128 129 130 131 132 133 134

/**
 * 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
135 136
 * @param num_groups number of channel groups
 * @param group_map  array with # of channels in group - 1, for each group
137 138 139
 *
 * @return zero if successful, a negative value if not
 */
140 141 142
int ff_psy_init(FFPsyContext *ctx, AVCodecContext *avctx, int num_lens,
                const uint8_t **bands, const int *num_bands,
                int num_groups, const uint8_t *group_map);
143 144 145 146 147 148 149 150 151 152

/**
 * Determine what group a channel belongs to.
 *
 * @param ctx     psymodel context
 * @param channel channel to locate the group for
 *
 * @return pointer to the FFPsyChannelGroup this channel belongs to
 */
FFPsyChannelGroup *ff_psy_find_group(FFPsyContext *ctx, int channel);
153 154 155 156 157 158

/**
 * Cleanup model context at the end.
 *
 * @param ctx model context
 */
159
void ff_psy_end(FFPsyContext *ctx);
160 161 162 163 164 165 166 167 168 169 170


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

/**
 * psychoacoustic model audio preprocessing initialization
 */
171
struct FFPsyPreprocessContext *ff_psy_preprocess_init(AVCodecContext *avctx);
172 173 174 175 176

/**
 * Preprocess several channel in audio frame in order to compress it better.
 *
 * @param ctx      preprocessing context
177 178
 * @param audio    samples to be filtered (in place)
 * @param channels number of channel to preprocess
179
 */
180
void ff_psy_preprocess(struct FFPsyPreprocessContext *ctx, float **audio, int channels);
181 182 183 184

/**
 * Cleanup audio preprocessing module.
 */
185
void ff_psy_preprocess_end(struct FFPsyPreprocessContext *ctx);
186 187

#endif /* AVCODEC_PSYMODEL_H */