2011-09-11 04:14:14 +03:00
|
|
|
/*
|
2011-10-18 22:16:27 +03:00
|
|
|
* Copyright (C) 2009 Justin Ruggles
|
|
|
|
* Copyright (c) 2009 Xuggle Incorporated
|
|
|
|
*
|
2011-09-11 04:14:14 +03:00
|
|
|
* 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
|
|
|
|
*/
|
2011-10-18 22:16:27 +03:00
|
|
|
|
|
|
|
/**
|
|
|
|
* @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->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 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.
|
|
|
|
*/
|
|
|
|
|
2011-09-11 04:14:14 +03:00
|
|
|
#include <speex/speex.h>
|
|
|
|
#include <speex/speex_header.h>
|
|
|
|
#include <speex/speex_stereo.h>
|
2011-10-18 22:16:27 +03:00
|
|
|
#include "libavutil/opt.h"
|
|
|
|
#include "avcodec.h"
|
|
|
|
#include "internal.h"
|
2012-02-28 04:01:42 +03:00
|
|
|
#include "audio_frame_queue.h"
|
2011-09-11 04:14:14 +03:00
|
|
|
|
|
|
|
typedef struct {
|
2011-10-18 22:16:27 +03:00
|
|
|
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 pkt_frame_count; ///< frame count for the current packet
|
2012-02-28 04:01:42 +03:00
|
|
|
AudioFrameQueue afq; ///< frame queue
|
2011-09-11 04:14:14 +03:00
|
|
|
} LibSpeexEncContext;
|
|
|
|
|
2011-10-18 22:16:27 +03:00
|
|
|
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->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: %d bps\n", avctx->bit_rate);
|
|
|
|
} else {
|
|
|
|
av_log(avctx, AV_LOG_DEBUG, "rate control: CBR\n");
|
|
|
|
av_log(avctx, AV_LOG_DEBUG, " bitrate: %d 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);
|
|
|
|
}
|
2011-09-11 04:14:14 +03:00
|
|
|
|
2011-10-18 22:16:27 +03:00
|
|
|
static av_cold int encode_init(AVCodecContext *avctx)
|
2011-09-11 04:14:14 +03:00
|
|
|
{
|
2011-10-18 22:16:27 +03:00
|
|
|
LibSpeexEncContext *s = avctx->priv_data;
|
2011-09-11 04:14:14 +03:00
|
|
|
const SpeexMode *mode;
|
2011-10-18 22:16:27 +03:00
|
|
|
uint8_t *header_data;
|
|
|
|
int header_size;
|
|
|
|
int32_t complexity;
|
2011-09-11 04:14:14 +03:00
|
|
|
|
2011-10-18 22:16:27 +03:00
|
|
|
/* channels */
|
|
|
|
if (avctx->channels < 1 || avctx->channels > 2) {
|
|
|
|
av_log(avctx, AV_LOG_ERROR, "Invalid channels (%d). Only stereo and "
|
|
|
|
"mono are supported\n", avctx->channels);
|
|
|
|
return AVERROR(EINVAL);
|
2011-09-11 04:14:14 +03:00
|
|
|
}
|
|
|
|
|
2011-10-18 22:16:27 +03:00
|
|
|
/* sample rate and encoding mode */
|
|
|
|
switch (avctx->sample_rate) {
|
|
|
|
case 8000: mode = &speex_nb_mode; break;
|
|
|
|
case 16000: mode = &speex_wb_mode; break;
|
|
|
|
case 32000: mode = &speex_uwb_mode; 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);
|
|
|
|
}
|
2011-09-11 04:14:14 +03:00
|
|
|
|
2011-10-18 22:16:27 +03:00
|
|
|
/* initialize libspeex */
|
2011-09-11 04:14:14 +03:00
|
|
|
s->enc_state = speex_encoder_init(mode);
|
2011-10-18 22:16:27 +03:00
|
|
|
if (!s->enc_state) {
|
|
|
|
av_log(avctx, AV_LOG_ERROR, "Error initializing libspeex\n");
|
2011-09-11 04:14:14 +03:00
|
|
|
return -1;
|
|
|
|
}
|
2011-10-18 22:16:27 +03:00
|
|
|
speex_init_header(&s->header, avctx->sample_rate, avctx->channels, mode);
|
2011-09-11 04:14:14 +03:00
|
|
|
|
2011-10-18 22:16:27 +03:00
|
|
|
/* rate control method and parameters */
|
2011-09-11 04:14:14 +03:00
|
|
|
if (avctx->flags & CODEC_FLAG_QSCALE) {
|
2011-10-18 22:16:27 +03:00
|
|
|
/* VBR */
|
|
|
|
s->header.vbr = 1;
|
|
|
|
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);
|
2011-09-11 04:14:14 +03:00
|
|
|
} else {
|
2011-10-18 22:16:27 +03:00
|
|
|
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 + (avctx->channels == 2 ? 800 : 0);
|
2011-09-11 04:14:14 +03:00
|
|
|
}
|
2011-10-18 22:16:27 +03:00
|
|
|
|
|
|
|
/* 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 */
|
2012-02-10 05:03:02 +03:00
|
|
|
speex_encoder_ctl(s->enc_state, SPEEX_GET_LOOKAHEAD, &avctx->delay);
|
2012-02-28 04:01:42 +03:00
|
|
|
ff_af_queue_init(avctx, &s->afq);
|
2011-10-18 22:16:27 +03:00
|
|
|
|
|
|
|
/* 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 and coded_frame */
|
|
|
|
avctx->extradata = av_malloc(header_size + FF_INPUT_BUFFER_PADDING_SIZE);
|
2012-02-28 04:01:42 +03:00
|
|
|
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);
|
|
|
|
}
|
|
|
|
#if FF_API_OLD_ENCODE_AUDIO
|
2011-10-18 22:16:27 +03:00
|
|
|
avctx->coded_frame = avcodec_alloc_frame();
|
2012-02-28 04:01:42 +03:00
|
|
|
if (!avctx->coded_frame) {
|
|
|
|
av_freep(&avctx->extradata);
|
2011-10-18 22:16:27 +03:00
|
|
|
speex_header_free(header_data);
|
|
|
|
speex_encoder_destroy(s->enc_state);
|
|
|
|
av_log(avctx, AV_LOG_ERROR, "memory allocation error\n");
|
|
|
|
return AVERROR(ENOMEM);
|
|
|
|
}
|
2012-02-28 04:01:42 +03:00
|
|
|
#endif
|
2011-10-18 22:16:27 +03:00
|
|
|
|
|
|
|
/* 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);
|
2011-09-11 04:14:14 +03:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2012-02-28 04:01:42 +03:00
|
|
|
static int encode_frame(AVCodecContext *avctx, AVPacket *avpkt,
|
|
|
|
const AVFrame *frame, int *got_packet_ptr)
|
2011-09-11 04:14:14 +03:00
|
|
|
{
|
2011-10-18 22:16:27 +03:00
|
|
|
LibSpeexEncContext *s = avctx->priv_data;
|
2012-02-28 04:01:42 +03:00
|
|
|
int16_t *samples = frame ? (int16_t *)frame->data[0] : NULL;
|
|
|
|
int ret;
|
2011-10-18 22:16:27 +03:00
|
|
|
|
2012-02-28 04:01:42 +03:00
|
|
|
if (samples) {
|
2011-10-18 22:16:27 +03:00
|
|
|
/* encode Speex frame */
|
|
|
|
if (avctx->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++;
|
2012-02-28 04:01:42 +03:00
|
|
|
if ((ret = ff_af_queue_add(&s->afq, frame) < 0))
|
|
|
|
return ret;
|
2011-10-18 22:16:27 +03:00
|
|
|
} 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++;
|
2011-09-11 04:14:14 +03:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2011-10-18 22:16:27 +03:00
|
|
|
/* 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;
|
2012-03-22 23:22:39 +03:00
|
|
|
if ((ret = ff_alloc_packet2(avctx, avpkt, speex_bits_nbytes(&s->bits))))
|
2011-10-18 22:16:27 +03:00
|
|
|
return ret;
|
2012-02-28 04:01:42 +03:00
|
|
|
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;
|
2011-09-11 04:14:14 +03:00
|
|
|
}
|
2011-10-18 22:16:27 +03:00
|
|
|
return 0;
|
2011-09-11 04:14:14 +03:00
|
|
|
}
|
|
|
|
|
2011-10-18 22:16:27 +03:00
|
|
|
static av_cold int encode_close(AVCodecContext *avctx)
|
2011-09-11 04:14:14 +03:00
|
|
|
{
|
2011-10-18 22:16:27 +03:00
|
|
|
LibSpeexEncContext *s = avctx->priv_data;
|
2011-09-11 04:14:14 +03:00
|
|
|
|
|
|
|
speex_bits_destroy(&s->bits);
|
|
|
|
speex_encoder_destroy(s->enc_state);
|
2011-10-18 22:16:27 +03:00
|
|
|
|
2012-02-28 04:01:42 +03:00
|
|
|
ff_af_queue_close(&s->afq);
|
|
|
|
#if FF_API_OLD_ENCODE_AUDIO
|
2011-10-18 22:16:27 +03:00
|
|
|
av_freep(&avctx->coded_frame);
|
2012-02-28 04:01:42 +03:00
|
|
|
#endif
|
2011-10-18 22:16:27 +03:00
|
|
|
av_freep(&avctx->extradata);
|
2011-09-11 04:14:14 +03:00
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2011-10-18 22:16:27 +03:00
|
|
|
#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, { 0 }, 0, 1, AE },
|
|
|
|
{ "cbr_quality", "Set quality value (0 to 10) for CBR", OFFSET(cbr_quality), AV_OPT_TYPE_INT, { 8 }, 0, 10, AE },
|
|
|
|
{ "frames_per_packet", "Number of frames to encode in each packet", OFFSET(frames_per_packet), AV_OPT_TYPE_INT, { 1 }, 1, 8, AE },
|
|
|
|
{ NULL },
|
|
|
|
};
|
|
|
|
|
|
|
|
static const AVClass class = {
|
|
|
|
.class_name = "libspeex",
|
|
|
|
.item_name = av_default_item_name,
|
|
|
|
.option = options,
|
|
|
|
.version = LIBAVUTIL_VERSION_INT,
|
|
|
|
};
|
|
|
|
|
|
|
|
static const AVCodecDefault defaults[] = {
|
|
|
|
{ "b", "0" },
|
|
|
|
{ "compression_level", "3" },
|
|
|
|
{ NULL },
|
|
|
|
};
|
|
|
|
|
2011-09-11 04:14:14 +03:00
|
|
|
AVCodec ff_libspeex_encoder = {
|
2011-10-18 22:16:27 +03:00
|
|
|
.name = "libspeex",
|
|
|
|
.type = AVMEDIA_TYPE_AUDIO,
|
|
|
|
.id = CODEC_ID_SPEEX,
|
|
|
|
.priv_data_size = sizeof(LibSpeexEncContext),
|
|
|
|
.init = encode_init,
|
2012-02-28 04:01:42 +03:00
|
|
|
.encode2 = encode_frame,
|
2011-10-18 22:16:27 +03:00
|
|
|
.close = encode_close,
|
|
|
|
.capabilities = CODEC_CAP_DELAY,
|
2012-04-06 19:19:39 +03:00
|
|
|
.sample_fmts = (const enum AVSampleFormat[]){ AV_SAMPLE_FMT_S16,
|
|
|
|
AV_SAMPLE_FMT_NONE },
|
2011-10-18 22:16:27 +03:00
|
|
|
.long_name = NULL_IF_CONFIG_SMALL("libspeex Speex"),
|
|
|
|
.priv_class = &class,
|
|
|
|
.defaults = defaults,
|
2011-09-11 04:14:14 +03:00
|
|
|
};
|