path: root/libavcodec/opus.h (plain)
blob: c3cbaec35d60b67937eecf1dfa82042708d9fc26

1	/*
2	* Opus decoder/demuxer common functions
3	* Copyright (c) 2012 Andrew D'Addesio
4	* Copyright (c) 2013-2014 Mozilla Corporation
5	*
6	* This file is part of FFmpeg.
7	*
8	* FFmpeg is free software; you can redistribute it and/or
9	* modify it under the terms of the GNU Lesser General Public
10	* License as published by the Free Software Foundation; either
11	* version 2.1 of the License, or (at your option) any later version.
12	*
13	* FFmpeg is distributed in the hope that it will be useful,
14	* but WITHOUT ANY WARRANTY; without even the implied warranty of
15	* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
16	* Lesser General Public License for more details.
17	*
18	* You should have received a copy of the GNU Lesser General Public
19	* License along with FFmpeg; if not, write to the Free Software
20	* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
21	*/
22
23	#ifndef AVCODEC_OPUS_H
24	#define AVCODEC_OPUS_H
25
26	#include <stdint.h>
27
28	#include "libavutil/audio_fifo.h"
29	#include "libavutil/float_dsp.h"
30	#include "libavutil/frame.h"
31
32	#include "libswresample/swresample.h"
33
34	#include "avcodec.h"
35	#include "opus_rc.h"
36
37	#define MAX_FRAME_SIZE 1275
38	#define MAX_FRAMES 48
39	#define MAX_PACKET_DUR 5760
40
41	#define CELT_SHORT_BLOCKSIZE 120
42	#define CELT_OVERLAP CELT_SHORT_BLOCKSIZE
43	#define CELT_MAX_LOG_BLOCKS 3
44	#define CELT_MAX_FRAME_SIZE (CELT_SHORT_BLOCKSIZE * (1 << CELT_MAX_LOG_BLOCKS))
45	#define CELT_MAX_BANDS 21
46
47	#define SILK_HISTORY 322
48	#define SILK_MAX_LPC 16
49
50	#define ROUND_MULL(a,b,s) (((MUL64(a, b) >> ((s) - 1)) + 1) >> 1)
51	#define ROUND_MUL16(a,b) ((MUL16(a, b) + 16384) >> 15)
52
53	#define OPUS_TS_HEADER 0x7FE0 // 0x3ff (11 bits)
54	#define OPUS_TS_MASK 0xFFE0 // top 11 bits
55
56	static const uint8_t opus_default_extradata[30] = {
57	'O', 'p', 'u', 's', 'H', 'e', 'a', 'd',
58	1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
59	0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
60	};
61
62	enum OpusMode {
63	OPUS_MODE_SILK,
64	OPUS_MODE_HYBRID,
65	OPUS_MODE_CELT,
66
67	OPUS_MODE_NB
68	};
69
70	enum OpusBandwidth {
71	OPUS_BANDWIDTH_NARROWBAND,
72	OPUS_BANDWIDTH_MEDIUMBAND,
73	OPUS_BANDWIDTH_WIDEBAND,
74	OPUS_BANDWIDTH_SUPERWIDEBAND,
75	OPUS_BANDWIDTH_FULLBAND,
76
77	OPUS_BANDWITH_NB
78	};
79
80	typedef struct SilkContext SilkContext;
81
82	typedef struct CeltFrame CeltFrame;
83
84	typedef struct OpusPacket {
85	int packet_size; /**< packet size */
86	int data_size; /**< size of the useful data -- packet size - padding */
87	int code; /**< packet code: specifies the frame layout */
88	int stereo; /**< whether this packet is mono or stereo */
89	int vbr; /**< vbr flag */
90	int config; /**< configuration: tells the audio mode,
91	** bandwidth, and frame duration */
92	int frame_count; /**< frame count */
93	int frame_offset[MAX_FRAMES]; /**< frame offsets */
94	int frame_size[MAX_FRAMES]; /**< frame sizes */
95	int frame_duration; /**< frame duration, in samples @ 48kHz */
96	enum OpusMode mode; /**< mode */
97	enum OpusBandwidth bandwidth; /**< bandwidth */
98	} OpusPacket;
99
100	typedef struct OpusStreamContext {
101	AVCodecContext *avctx;
102	int output_channels;
103
104	OpusRangeCoder rc;
105	OpusRangeCoder redundancy_rc;
106	SilkContext *silk;
107	CeltFrame *celt;
108	AVFloatDSPContext *fdsp;
109
110	float silk_buf[2][960];
111	float *silk_output[2];
112	DECLARE_ALIGNED(32, float, celt_buf)[2][960];
113	float *celt_output[2];
114
115	float redundancy_buf[2][960];
116	float *redundancy_output[2];
117
118	/* data buffers for the final output data */
119	float *out[2];
120	int out_size;
121
122	float *out_dummy;
123	int out_dummy_allocated_size;
124
125	SwrContext *swr;
126	AVAudioFifo *celt_delay;
127	int silk_samplerate;
128	/* number of samples we still want to get from the resampler */
129	int delayed_samples;
130
131	OpusPacket packet;
132
133	int redundancy_idx;
134	} OpusStreamContext;
135
136	// a mapping between an opus stream and an output channel
137	typedef struct ChannelMap {
138	int stream_idx;
139	int channel_idx;
140
141	// when a single decoded channel is mapped to multiple output channels, we
142	// write to the first output directly and copy from it to the others
143	// this field is set to 1 for those copied output channels
144	int copy;
145	// this is the index of the output channel to copy from
146	int copy_idx;
147
148	// this channel is silent
149	int silence;
150	} ChannelMap;
151
152	typedef struct OpusContext {
153	OpusStreamContext *streams;
154
155	/* current output buffers for each streams */
156	float **out;
157	int *out_size;
158	/* Buffers for synchronizing the streams when they have different
159	* resampling delays */
160	AVAudioFifo **sync_buffers;
161	/* number of decoded samples for each stream */
162	int *decoded_samples;
163
164	int nb_streams;
165	int nb_stereo_streams;
166
167	AVFloatDSPContext *fdsp;
168	int16_t gain_i;
169	float gain;
170
171	ChannelMap *channel_maps;
172	} OpusContext;
173
174	int ff_opus_parse_packet(OpusPacket *pkt, const uint8_t *buf, int buf_size,
175	int self_delimited);
176
177	int ff_opus_parse_extradata(AVCodecContext *avctx, OpusContext *s);
178
179	int ff_silk_init(AVCodecContext *avctx, SilkContext **ps, int output_channels);
180	void ff_silk_free(SilkContext **ps);
181	void ff_silk_flush(SilkContext *s);
182
183	/**
184	* Decode the LP layer of one Opus frame (which may correspond to several SILK
185	* frames).
186	*/
187	int ff_silk_decode_superframe(SilkContext *s, OpusRangeCoder *rc,
188	float *output[2],
189	enum OpusBandwidth bandwidth, int coded_channels,
190	int duration_ms);
191
192	#endif /* AVCODEC_OPUS_H */
193