1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362
|
/*
* Copyright (C) 2009 Justin Ruggles
* Copyright (c) 2009 Xuggle Incorporated
*
* 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
*/
/**
* @file
* libspeex Speex audio encoder
*
* Usage Guide
* This explains the values that need to be set prior to initialization in
* order to control various encoding parameters.
*
* Channels
* Speex only supports mono or stereo, so avctx->ch_layout.nb_channels must
* be set to 1 or 2.
*
* Sample Rate / Encoding Mode
* Speex has 3 modes, each of which uses a specific sample rate.
* narrowband : 8 kHz
* wideband : 16 kHz
* ultra-wideband : 32 kHz
* avctx->sample_rate must be set to one of these 3 values. This will be
* used to set the encoding mode.
*
* Rate Control
* VBR mode is turned on by setting AV_CODEC_FLAG_QSCALE in avctx->flags.
* avctx->global_quality is used to set the encoding quality.
* For CBR mode, avctx->bit_rate can be used to set the constant bitrate.
* Alternatively, the 'cbr_quality' option can be set from 0 to 10 to set
* a constant bitrate based on quality.
* For ABR mode, set avctx->bit_rate and set the 'abr' option to 1.
* Approx. Bitrate Range:
* narrowband : 2400 - 25600 bps
* wideband : 4000 - 43200 bps
* ultra-wideband : 4400 - 45200 bps
*
* Complexity
* Encoding complexity is controlled by setting avctx->compression_level.
* The valid range is 0 to 10. A higher setting gives generally better
* quality at the expense of encoding speed. This does not affect the
* bit rate.
*
* Frames-per-Packet
* The encoder defaults to using 1 frame-per-packet. However, it is
* sometimes desirable to use multiple frames-per-packet to reduce the
* amount of container overhead. This can be done by setting the
* 'frames_per_packet' option to a value 1 to 8.
*
*
* Optional features
* Speex encoder supports several optional features, which can be useful
* for some conditions.
*
* Voice Activity Detection
* When enabled, voice activity detection detects whether the audio
* being encoded is speech or silence/background noise. VAD is always
* implicitly activated when encoding in VBR, so the option is only useful
* in non-VBR operation. In this case, Speex detects non-speech periods and
* encodes them with just enough bits to reproduce the background noise.
*
* Discontinuous Transmission (DTX)
* DTX is an addition to VAD/VBR operation, that makes it possible to stop transmitting
* completely when the background noise is stationary.
* In file-based operation only 5 bits are used for such frames.
*/
#include <speex/speex.h>
#include <speex/speex_header.h>
#include <speex/speex_stereo.h>
#include "libavutil/channel_layout.h"
#include "libavutil/common.h"
#include "libavutil/mem.h"
#include "libavutil/opt.h"
#include "avcodec.h"
#include "codec_internal.h"
#include "encode.h"
#include "audio_frame_queue.h"
/* TODO: Think about converting abr, vad, dtx and such flags to a bit field */
typedef struct LibSpeexEncContext {
AVClass *class; ///< AVClass for private options
SpeexBits bits; ///< libspeex bitwriter context
SpeexHeader header; ///< libspeex header struct
void *enc_state; ///< libspeex encoder state
int frames_per_packet; ///< number of frames to encode in each packet
float vbr_quality; ///< VBR quality 0.0 to 10.0
int cbr_quality; ///< CBR quality 0 to 10
int abr; ///< flag to enable ABR
int vad; ///< flag to enable VAD
int dtx; ///< flag to enable DTX
int pkt_frame_count; ///< frame count for the current packet
AudioFrameQueue afq; ///< frame queue
} LibSpeexEncContext;
static av_cold void print_enc_params(AVCodecContext *avctx,
LibSpeexEncContext *s)
{
const char *mode_str = "unknown";
av_log(avctx, AV_LOG_DEBUG, "channels: %d\n", avctx->ch_layout.nb_channels);
switch (s->header.mode) {
case SPEEX_MODEID_NB: mode_str = "narrowband"; break;
case SPEEX_MODEID_WB: mode_str = "wideband"; break;
case SPEEX_MODEID_UWB: mode_str = "ultra-wideband"; break;
}
av_log(avctx, AV_LOG_DEBUG, "mode: %s\n", mode_str);
if (s->header.vbr) {
av_log(avctx, AV_LOG_DEBUG, "rate control: VBR\n");
av_log(avctx, AV_LOG_DEBUG, " quality: %f\n", s->vbr_quality);
} else if (s->abr) {
av_log(avctx, AV_LOG_DEBUG, "rate control: ABR\n");
av_log(avctx, AV_LOG_DEBUG, " bitrate: %"PRId64" bps\n", avctx->bit_rate);
} else {
av_log(avctx, AV_LOG_DEBUG, "rate control: CBR\n");
av_log(avctx, AV_LOG_DEBUG, " bitrate: %"PRId64" bps\n", avctx->bit_rate);
}
av_log(avctx, AV_LOG_DEBUG, "complexity: %d\n",
avctx->compression_level);
av_log(avctx, AV_LOG_DEBUG, "frame size: %d samples\n",
avctx->frame_size);
av_log(avctx, AV_LOG_DEBUG, "frames per packet: %d\n",
s->frames_per_packet);
av_log(avctx, AV_LOG_DEBUG, "packet size: %d\n",
avctx->frame_size * s->frames_per_packet);
av_log(avctx, AV_LOG_DEBUG, "voice activity detection: %d\n", s->vad);
av_log(avctx, AV_LOG_DEBUG, "discontinuous transmission: %d\n", s->dtx);
}
static av_cold int encode_init(AVCodecContext *avctx)
{
LibSpeexEncContext *s = avctx->priv_data;
int channels = avctx->ch_layout.nb_channels;
const SpeexMode *mode;
uint8_t *header_data;
int header_size;
int32_t complexity;
/* sample rate and encoding mode */
switch (avctx->sample_rate) {
case 8000: mode = speex_lib_get_mode(SPEEX_MODEID_NB); break;
case 16000: mode = speex_lib_get_mode(SPEEX_MODEID_WB); break;
case 32000: mode = speex_lib_get_mode(SPEEX_MODEID_UWB); break;
default:
av_log(avctx, AV_LOG_ERROR, "Sample rate of %d Hz is not supported. "
"Resample to 8, 16, or 32 kHz.\n", avctx->sample_rate);
return AVERROR(EINVAL);
}
/* initialize libspeex */
s->enc_state = speex_encoder_init(mode);
if (!s->enc_state) {
av_log(avctx, AV_LOG_ERROR, "Error initializing libspeex\n");
return -1;
}
speex_init_header(&s->header, avctx->sample_rate, channels, mode);
/* rate control method and parameters */
if (avctx->flags & AV_CODEC_FLAG_QSCALE) {
/* VBR */
s->header.vbr = 1;
s->vad = 1; /* VAD is always implicitly activated for VBR */
speex_encoder_ctl(s->enc_state, SPEEX_SET_VBR, &s->header.vbr);
s->vbr_quality = av_clipf(avctx->global_quality / (float)FF_QP2LAMBDA,
0.0f, 10.0f);
speex_encoder_ctl(s->enc_state, SPEEX_SET_VBR_QUALITY, &s->vbr_quality);
} else {
s->header.bitrate = avctx->bit_rate;
if (avctx->bit_rate > 0) {
/* CBR or ABR by bitrate */
if (s->abr) {
speex_encoder_ctl(s->enc_state, SPEEX_SET_ABR,
&s->header.bitrate);
speex_encoder_ctl(s->enc_state, SPEEX_GET_ABR,
&s->header.bitrate);
} else {
speex_encoder_ctl(s->enc_state, SPEEX_SET_BITRATE,
&s->header.bitrate);
speex_encoder_ctl(s->enc_state, SPEEX_GET_BITRATE,
&s->header.bitrate);
}
} else {
/* CBR by quality */
speex_encoder_ctl(s->enc_state, SPEEX_SET_QUALITY,
&s->cbr_quality);
speex_encoder_ctl(s->enc_state, SPEEX_GET_BITRATE,
&s->header.bitrate);
}
/* stereo side information adds about 800 bps to the base bitrate */
/* TODO: this should be calculated exactly */
avctx->bit_rate = s->header.bitrate + (channels == 2 ? 800 : 0);
}
/* VAD is activated with VBR or can be turned on by itself */
if (s->vad)
speex_encoder_ctl(s->enc_state, SPEEX_SET_VAD, &s->vad);
/* Activating Discontinuous Transmission */
if (s->dtx) {
speex_encoder_ctl(s->enc_state, SPEEX_SET_DTX, &s->dtx);
if (!(s->abr || s->vad || s->header.vbr))
av_log(avctx, AV_LOG_WARNING, "DTX is not much of use without ABR, VAD or VBR\n");
}
/* set encoding complexity */
if (avctx->compression_level > FF_COMPRESSION_DEFAULT) {
complexity = av_clip(avctx->compression_level, 0, 10);
speex_encoder_ctl(s->enc_state, SPEEX_SET_COMPLEXITY, &complexity);
}
speex_encoder_ctl(s->enc_state, SPEEX_GET_COMPLEXITY, &complexity);
avctx->compression_level = complexity;
/* set packet size */
avctx->frame_size = s->header.frame_size;
s->header.frames_per_packet = s->frames_per_packet;
/* set encoding delay */
speex_encoder_ctl(s->enc_state, SPEEX_GET_LOOKAHEAD, &avctx->initial_padding);
ff_af_queue_init(avctx, &s->afq);
/* create header packet bytes from header struct */
/* note: libspeex allocates the memory for header_data, which is freed
below with speex_header_free() */
header_data = speex_header_to_packet(&s->header, &header_size);
/* allocate extradata */
avctx->extradata = av_malloc(header_size + AV_INPUT_BUFFER_PADDING_SIZE);
if (!avctx->extradata) {
speex_header_free(header_data);
speex_encoder_destroy(s->enc_state);
av_log(avctx, AV_LOG_ERROR, "memory allocation error\n");
return AVERROR(ENOMEM);
}
/* copy header packet to extradata */
memcpy(avctx->extradata, header_data, header_size);
avctx->extradata_size = header_size;
speex_header_free(header_data);
/* init libspeex bitwriter */
speex_bits_init(&s->bits);
print_enc_params(avctx, s);
return 0;
}
static int encode_frame(AVCodecContext *avctx, AVPacket *avpkt,
const AVFrame *frame, int *got_packet_ptr)
{
LibSpeexEncContext *s = avctx->priv_data;
int16_t *samples = frame ? (int16_t *)frame->data[0] : NULL;
int ret;
if (samples) {
/* encode Speex frame */
if (avctx->ch_layout.nb_channels == 2)
speex_encode_stereo_int(samples, s->header.frame_size, &s->bits);
speex_encode_int(s->enc_state, samples, &s->bits);
s->pkt_frame_count++;
if ((ret = ff_af_queue_add(&s->afq, frame)) < 0)
return ret;
} else {
/* handle end-of-stream */
if (!s->pkt_frame_count)
return 0;
/* add extra terminator codes for unused frames in last packet */
while (s->pkt_frame_count < s->frames_per_packet) {
speex_bits_pack(&s->bits, 15, 5);
s->pkt_frame_count++;
}
}
/* write output if all frames for the packet have been encoded */
if (s->pkt_frame_count == s->frames_per_packet) {
s->pkt_frame_count = 0;
if ((ret = ff_alloc_packet(avctx, avpkt, speex_bits_nbytes(&s->bits))) < 0)
return ret;
ret = speex_bits_write(&s->bits, avpkt->data, avpkt->size);
speex_bits_reset(&s->bits);
/* Get the next frame pts/duration */
ff_af_queue_remove(&s->afq, s->frames_per_packet * avctx->frame_size,
&avpkt->pts, &avpkt->duration);
avpkt->size = ret;
*got_packet_ptr = 1;
return 0;
}
return 0;
}
static av_cold int encode_close(AVCodecContext *avctx)
{
LibSpeexEncContext *s = avctx->priv_data;
speex_bits_destroy(&s->bits);
speex_encoder_destroy(s->enc_state);
ff_af_queue_close(&s->afq);
return 0;
}
#define OFFSET(x) offsetof(LibSpeexEncContext, x)
#define AE AV_OPT_FLAG_AUDIO_PARAM | AV_OPT_FLAG_ENCODING_PARAM
static const AVOption options[] = {
{ "abr", "Use average bit rate", OFFSET(abr), AV_OPT_TYPE_INT, { .i64 = 0 }, 0, 1, AE },
{ "cbr_quality", "Set quality value (0 to 10) for CBR", OFFSET(cbr_quality), AV_OPT_TYPE_INT, { .i64 = 8 }, 0, 10, AE },
{ "frames_per_packet", "Number of frames to encode in each packet", OFFSET(frames_per_packet), AV_OPT_TYPE_INT, { .i64 = 1 }, 1, 8, AE },
{ "vad", "Voice Activity Detection", OFFSET(vad), AV_OPT_TYPE_INT, { .i64 = 0 }, 0, 1, AE },
{ "dtx", "Discontinuous Transmission", OFFSET(dtx), AV_OPT_TYPE_INT, { .i64 = 0 }, 0, 1, AE },
{ NULL },
};
static const AVClass speex_class = {
.class_name = "libspeex",
.item_name = av_default_item_name,
.option = options,
.version = LIBAVUTIL_VERSION_INT,
};
static const FFCodecDefault defaults[] = {
{ "b", "0" },
{ "compression_level", "3" },
{ NULL },
};
const FFCodec ff_libspeex_encoder = {
.p.name = "libspeex",
CODEC_LONG_NAME("libspeex Speex"),
.p.type = AVMEDIA_TYPE_AUDIO,
.p.id = AV_CODEC_ID_SPEEX,
.p.capabilities = AV_CODEC_CAP_DR1 | AV_CODEC_CAP_DELAY,
.caps_internal = FF_CODEC_CAP_NOT_INIT_THREADSAFE,
.priv_data_size = sizeof(LibSpeexEncContext),
.init = encode_init,
FF_CODEC_ENCODE_CB(encode_frame),
.close = encode_close,
CODEC_SAMPLEFMTS(AV_SAMPLE_FMT_S16),
CODEC_CH_LAYOUTS(AV_CHANNEL_LAYOUT_MONO, AV_CHANNEL_LAYOUT_STEREO),
CODEC_SAMPLERATES(8000, 16000, 32000),
.p.priv_class = &speex_class,
.defaults = defaults,
.p.wrapper_name = "libspeex",
};
|