1
0
mirror of https://github.com/FFmpeg/FFmpeg.git synced 2024-12-23 12:43:46 +02:00
FFmpeg/libavcodec/opus_celt.h
Rostislav Pehlivanov 5f47c85e5c opus: add a native Opus encoder
This marks the first time anyone has written an Opus encoder without
using any libopus code. The aim of the encoder is to prove how far
the format can go by writing the craziest encoder for it.

Right now the encoder's basic, it only supports CBR encoding, however
internally every single feature the CELT layer has is implemented
(except the pitch pre-filter which needs to work well with the rest of
whatever gets implemented). Psychoacoustic and rate control systems are
under development.

The encoder takes in frames of 120 samples and depending on the value of
opus_delay the plan is to use the extra buffered frames as lookahead.
Right now the encoder will pick the nearest largest legal frame size and
won't use the lookahead, but that'll change once there's a
psychoacoustic system.

Even though its a pretty basic encoder its already outperforming
any other native encoder FFmpeg has by a huge amount.

The PVQ search algorithm is faster and more accurate than libopus's
algorithm so the encoder's performance is close to that of libopus
at zero complexity (libopus has more SIMD).
The algorithm might be ported to libopus or other codecs using PVQ in
the future.

The encoder still has a few minor bugs, like desyncs at ultra low
bitrates (below 9kbps with 20ms frames).

Signed-off-by: Rostislav Pehlivanov <atomnuker@gmail.com>
2017-02-14 06:15:36 +00:00

165 lines
4.6 KiB
C

/*
* Opus decoder/demuxer common functions
* Copyright (c) 2012 Andrew D'Addesio
* Copyright (c) 2013-2014 Mozilla Corporation
* Copyright (c) 2016 Rostislav Pehlivanov <atomnuker@gmail.com>
*
* This file is part of FFmpeg.
*
* FFmpeg is free software; you can redistribute it and/or
* 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.
*
* FFmpeg is distributed in the hope that it will be useful,
* 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
* License along with FFmpeg; if not, write to the Free Software
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
*/
#ifndef AVCODEC_OPUS_CELT_H
#define AVCODEC_OPUS_CELT_H
#include <float.h>
#include "opus.h"
#include "mdct15.h"
#include "libavutil/float_dsp.h"
#include "libavutil/libm.h"
#define CELT_VECTORS 11
#define CELT_ALLOC_STEPS 6
#define CELT_FINE_OFFSET 21
#define CELT_MAX_FINE_BITS 8
#define CELT_NORM_SCALE 16384
#define CELT_QTHETA_OFFSET 4
#define CELT_QTHETA_OFFSET_TWOPHASE 16
#define CELT_EMPH_COEFF 0.85000610f
#define CELT_POSTFILTER_MINPERIOD 15
#define CELT_ENERGY_SILENCE (-28.0f)
enum CeltSpread {
CELT_SPREAD_NONE,
CELT_SPREAD_LIGHT,
CELT_SPREAD_NORMAL,
CELT_SPREAD_AGGRESSIVE
};
enum CeltBlockSize {
CELT_BLOCK_120,
CELT_BLOCK_240,
CELT_BLOCK_480,
CELT_BLOCK_960,
CELT_BLOCK_NB
};
typedef struct CeltBlock {
float energy[CELT_MAX_BANDS];
float lin_energy[CELT_MAX_BANDS];
float error_energy[CELT_MAX_BANDS];
float prev_energy[2][CELT_MAX_BANDS];
uint8_t collapse_masks[CELT_MAX_BANDS];
int band_bins[CELT_MAX_BANDS]; /* MDCT bins per band */
float *band_coeffs[CELT_MAX_BANDS];
/* buffer for mdct output + postfilter */
DECLARE_ALIGNED(32, float, buf)[2048];
DECLARE_ALIGNED(32, float, coeffs)[CELT_MAX_FRAME_SIZE];
/* Used by the encoder */
DECLARE_ALIGNED(32, float, overlap)[120];
DECLARE_ALIGNED(32, float, samples)[CELT_MAX_FRAME_SIZE];
/* postfilter parameters */
int pf_period_new;
float pf_gains_new[3];
int pf_period;
float pf_gains[3];
int pf_period_old;
float pf_gains_old[3];
float emph_coeff;
} CeltBlock;
struct CeltFrame {
// constant values that do not change during context lifetime
AVCodecContext *avctx;
MDCT15Context *imdct[4];
AVFloatDSPContext *dsp;
CeltBlock block[2];
int channels;
int output_channels;
enum CeltBlockSize size;
int start_band;
int end_band;
int coded_bands;
int transient;
int intra;
int pfilter;
int skip_band_floor;
int tf_select;
int alloc_trim;
int alloc_boost[CELT_MAX_BANDS];
int blocks; /* number of iMDCT blocks in the frame, depends on transient */
int blocksize; /* size of each block */
int silence; /* Frame is filled with silence */
int anticollapse_needed; /* Whether to expect an anticollapse bit */
int anticollapse; /* Encoded anticollapse bit */
int intensity_stereo;
int dual_stereo;
int flushed;
uint32_t seed;
enum CeltSpread spread;
/* Bit allocation */
int framebits;
int remaining;
int remaining2;
int caps [CELT_MAX_BANDS];
int fine_bits [CELT_MAX_BANDS];
int fine_priority[CELT_MAX_BANDS];
int pulses [CELT_MAX_BANDS];
int tf_change [CELT_MAX_BANDS];
DECLARE_ALIGNED(32, float, scratch)[22 * 8]; // MAX(ff_celt_freq_range) * 1<<CELT_MAX_LOG_BLOCKS
};
/* LCG for noise generation */
static av_always_inline uint32_t celt_rng(CeltFrame *f)
{
f->seed = 1664525 * f->seed + 1013904223;
return f->seed;
}
static av_always_inline void celt_renormalize_vector(float *X, int N, float gain)
{
int i;
float g = 1e-15f;
for (i = 0; i < N; i++)
g += X[i] * X[i];
g = gain / sqrtf(g);
for (i = 0; i < N; i++)
X[i] *= g;
}
int ff_celt_init(AVCodecContext *avctx, CeltFrame **f, int output_channels);
void ff_celt_free(CeltFrame **f);
void ff_celt_flush(CeltFrame *f);
int ff_celt_decode_frame(CeltFrame *f, OpusRangeCoder *rc, float **output,
int coded_channels, int frame_size, int startband, int endband);
#endif /* AVCODEC_OPUS_CELT_H */