michael@0: /* Copyright (c) 2011 Xiph.Org Foundation
michael@0:    Written by Jean-Marc Valin */
michael@0: /*
michael@0:    Redistribution and use in source and binary forms, with or without
michael@0:    modification, are permitted provided that the following conditions
michael@0:    are met:
michael@0: 
michael@0:    - Redistributions of source code must retain the above copyright
michael@0:    notice, this list of conditions and the following disclaimer.
michael@0: 
michael@0:    - Redistributions in binary form must reproduce the above copyright
michael@0:    notice, this list of conditions and the following disclaimer in the
michael@0:    documentation and/or other materials provided with the distribution.
michael@0: 
michael@0:    THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
michael@0:    ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
michael@0:    LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
michael@0:    A PARTICULAR PURPOSE ARE DISCLAIMED.  IN NO EVENT SHALL THE FOUNDATION OR
michael@0:    CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
michael@0:    EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
michael@0:    PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
michael@0:    PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF
michael@0:    LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
michael@0:    NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
michael@0:    SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
michael@0: */
michael@0: 
michael@0: #ifndef ANALYSIS_H
michael@0: #define ANALYSIS_H
michael@0: 
michael@0: #include "celt.h"
michael@0: #include "opus_private.h"
michael@0: 
michael@0: #define NB_FRAMES 8
michael@0: #define NB_TBANDS 18
michael@0: #define NB_TOT_BANDS 21
michael@0: #define ANALYSIS_BUF_SIZE 720 /* 15 ms at 48 kHz */
michael@0: 
michael@0: #define DETECT_SIZE 200
michael@0: 
michael@0: typedef struct {
michael@0:    float angle[240];
michael@0:    float d_angle[240];
michael@0:    float d2_angle[240];
michael@0:    opus_val32 inmem[ANALYSIS_BUF_SIZE];
michael@0:    int   mem_fill;                      /* number of usable samples in the buffer */
michael@0:    float prev_band_tonality[NB_TBANDS];
michael@0:    float prev_tonality;
michael@0:    float E[NB_FRAMES][NB_TBANDS];
michael@0:    float lowE[NB_TBANDS];
michael@0:    float highE[NB_TBANDS];
michael@0:    float meanE[NB_TOT_BANDS];
michael@0:    float mem[32];
michael@0:    float cmean[8];
michael@0:    float std[9];
michael@0:    float music_prob;
michael@0:    float Etracker;
michael@0:    float lowECount;
michael@0:    int E_count;
michael@0:    int last_music;
michael@0:    int last_transition;
michael@0:    int count;
michael@0:    float subframe_mem[3];
michael@0:    int analysis_offset;
michael@0:    /** Probability of having speech for time i to DETECT_SIZE-1 (and music before).
michael@0:        pspeech[0] is the probability that all frames in the window are speech. */
michael@0:    float pspeech[DETECT_SIZE];
michael@0:    /** Probability of having music for time i to DETECT_SIZE-1 (and speech before).
michael@0:        pmusic[0] is the probability that all frames in the window are music. */
michael@0:    float pmusic[DETECT_SIZE];
michael@0:    float speech_confidence;
michael@0:    float music_confidence;
michael@0:    int speech_confidence_count;
michael@0:    int music_confidence_count;
michael@0:    int write_pos;
michael@0:    int read_pos;
michael@0:    int read_subframe;
michael@0:    AnalysisInfo info[DETECT_SIZE];
michael@0: } TonalityAnalysisState;
michael@0: 
michael@0: void tonality_analysis(TonalityAnalysisState *tonal, AnalysisInfo *info,
michael@0:      const CELTMode *celt_mode, const void *x, int len, int offset, int c1, int c2, int C, int lsb_depth, downmix_func downmix);
michael@0: 
michael@0: void tonality_get_info(TonalityAnalysisState *tonal, AnalysisInfo *info_out, int len);
michael@0: 
michael@0: void run_analysis(TonalityAnalysisState *analysis, const CELTMode *celt_mode, const void *analysis_pcm,
michael@0:                  int analysis_frame_size, int frame_size, int c1, int c2, int C, opus_int32 Fs,
michael@0:                  int lsb_depth, downmix_func downmix, AnalysisInfo *analysis_info);
michael@0: 
michael@0: #endif