AAC的音频编码和解码实现
折腾了几天的AAC编码和解码，最开始用的是ffmpeg的接口，实现好实现，但是调试总是有各种问题，最后还是使用faac-1.28/faad2-2.7实现了AAC的编码和解码功能，使用这两个库的原因，是因为手里有另一套代码工程，已经实现了AAC的编码和解码，所以就直接拿来用了。代码参考：https://gitee.com/anyRTC/anyRTC-RTMP-OpenSource/blob/master/ArLiveLite/codec/aacencode.cc
libfaad2-2.7
默认PS_DEC开着，所以单声道又被修改为了双声道，出现解码错误！
#if (defined(PS_DEC) || defined(DRM_PS))
/* check if we have a mono file */
if (*channels == 1)
{
/* upMatrix to 2 channels for implicit signalling of PS */
*channels = 2;
}
#endif使用FFMPEG实现的aac编解码程序，没有实际调通，代码仅供参考！
问题包括：
1、内置的aac-codec不支持AV_SAMPLE_FMT_S16；
2、av_frame_get_buffer可能会报错；
//aac_codec.h #ifndef __AAC_CODEC_H__ #define __AAC_CODEC_H__ extern int close_aac_codec(void); extern int aac_decode_frame(unsigned char* inbuf, unsigned int inlen, unsigned char* outbuf, unsigned int* outlen); extern int aac_encode_frame(unsigned char* inbuf, unsigned int inlen, unsigned char* outbuf, unsigned int* outlen); extern int init_aac_codec(uint8_t encode_flag, int sample_rate, int channels, int bitrate); #endif
//aac_codec.cpp
//aac_codec.cpp
#ifdef __cplusplus
extern "C" {
#endif
#include "libavutil/time.h"
#include "libavformat/avformat.h"
#include "libavcodec/avcodec.h"
#include "libavdevice/avdevice.h"
#include "libswresample/swresample.h"
#include "libswscale/swscale.h"
#ifdef __cplusplus
}
#endif
#include "BroadcastJson.h"
#include "logutils.h"
#include "aac_codec.h"
/* Global timestamp for the audio frames. */
static int64_t pts = 0;
static AVCodecContext *input_codec_context = NULL;
static AVFrame *input_codec_frame = NULL;
static void init_packet(AVPacket *packet)
{
av_init_packet(packet);
/* Set the packet data and size so that it is recognized as being empty. */
packet->data = NULL;
packet->size = 0;
}
/*
sample_rate:16000
channels:1
bitrate:64000
*/
int init_aac_codec(uint8_t encode_flag, int sample_rate, int channels, int bitrate){
AVCodec *input_codec;
AVCodecContext *avctx;
int error;
AVCodecID codec_id = AV_CODEC_ID_AAC;
/* Find a decoder for the audio stream. */
if (encode_flag){
input_codec = avcodec_find_encoder_by_name("libfdk_aac");
//input_codec = avcodec_find_encoder(codec_id);
if (!input_codec) {
LOGD("Could not find input encodec\n");
return -1;
}
}else{
input_codec = avcodec_find_decoder_by_name("libfdk_aac");
//input_codec = avcodec_find_decoder(codec_id);
if (!input_codec) {
LOGD("Could not find input decodec\n");
return -1;
}
}
//first close aaccodec.
close_aac_codec();
/* Allocate a new decoding context. */
avctx = avcodec_alloc_context3(input_codec);
if (!avctx) {
LOGD("Could not allocate a decoding context\n");
return -1;
}
avctx->strict_std_compliance =FF_COMPLIANCE_EXPERIMENTAL;
//avctx->codec_id = codec_id;
avctx->sample_fmt = AV_SAMPLE_FMT_S16;//AV_SAMPLE_FMT_FLTP
avctx->sample_rate = sample_rate;
avctx->channels = channels;
avctx->channel_layout = av_get_default_channel_layout(channels);
// 音频编码规格
avctx->profile = FF_PROFILE_AAC_HE_V2;
avctx->bit_rate = bitrate;
/* Open the decoder for the audio stream to use it later. */
if ((error = avcodec_open2(avctx, input_codec, NULL)) < 0) {
LOGD("Could not open input codec (error '%d')\n", error);
avcodec_free_context(&avctx);
return error;
}
/* frame containing input raw audio */
input_codec_frame = av_frame_alloc();
if (!input_codec_frame) {
LOGD("Could not allocate audio frame\n");
return error;
}
input_codec_frame->sample_rate = sample_rate;
input_codec_frame->channels = channels;
if (encode_flag){
input_codec_frame->nb_samples = avctx->frame_size;//(sample_rate*channels*2*20)/8000;//
}else{
input_codec_frame->nb_samples = (sample_rate/1000)*20*channels;//avctx->frame_size;//(sample_rate*channels*2*20)/8000;//
}
input_codec_frame->format = avctx->sample_fmt;
input_codec_frame->channel_layout = avctx->channel_layout;
/* allocate the data buffers */
int ret = av_frame_get_buffer(input_codec_frame, 0);
if (ret < 0) {
LOGD("Could not allocate audio data buffers ret:%d avctx->frame_size:%d\n", ret, avctx->frame_size);
avcodec_free_context(&avctx);
return error;
}
/* Save the decoder context for easier access later. */
input_codec_context = avctx;
LOGD("init_aac_codec init sample_rate:%d, channels:%d, bitrate:%d.", sample_rate, channels, bitrate);
return 0;
}
int aac_encode_frame(unsigned char* inbuf, unsigned int inlen, unsigned char* outbuf, unsigned int* outlen)
{
//
AVPacket output_packet;
uint16_t *samples ;
int data_present = 0;
int index = 0;
int ret = 0;
init_packet(&output_packet);
if (input_codec_frame == NULL){
return -1;
}
/* make sure the frame is writable -- makes a copy if the encoder
* kept a reference internally */
ret = av_frame_make_writable(input_codec_frame);
samples = (uint16_t*)input_codec_frame->data[0];
memcpy((void *)samples, (void*)inbuf, inlen);
/* Set a timestamp based on the sample rate for the container. */
if (input_codec_frame) {
input_codec_frame->pts = pts;
pts += input_codec_frame->nb_samples;
}
/* send the packet with the compressed data to the decoder */
ret = avcodec_send_frame(input_codec_context, input_codec_frame);
if (ret < 0) {
LOGD("Error submitting the packet to the decoder\n");
exit(1);
}
/* read all the output frames (in general there may be any number of them */
while (ret >= 0) {
ret = avcodec_receive_packet(input_codec_context, &output_packet);
if (ret == AVERROR(EAGAIN) || ret == AVERROR_EOF)
break;
else if (ret < 0) {
LOGD("Error during decoding\n");
exit(1);
}
int data_size = av_get_bytes_per_sample(input_codec_context->sample_fmt);
if (data_size < 0) {
/* This should not occur, checking just for paranoia */
LOGD("Failed to calculate data size\n");
return -1;
}
memcpy(outbuf+index, output_packet.data, output_packet.size);
index+=output_packet.size;
}
*outlen = index;
return 0;
}
int aac_decode_frame(unsigned char* inbuf, unsigned int inlen, unsigned char* outbuf, unsigned int* outlen)
{
AVPacket output_packet;
uint16_t *samples ;
int data_present = 0;
int index = 0;
int ret = 0;
if (input_codec_frame == NULL){
return -1;
}
init_packet(&output_packet);
/* make sure the frame is writable -- makes a copy if the encoder
* kept a reference internally */
ret = av_frame_make_writable(input_codec_frame);
samples = (uint16_t*)input_codec_frame->data[0];
memcpy((void *)samples, (void*)inbuf, inlen);
/* send the packet with the compressed data to the decoder */
ret = avcodec_send_frame(input_codec_context, input_codec_frame);
if (ret < 0) {
LOGD("Error submitting the packet to the decoder, inlen:%d\n", inlen);
return -1;
}
/* read all the output frames (in general there may be any number of them */
while (ret >= 0) {
ret = avcodec_receive_packet(input_codec_context, &output_packet);
if (ret == AVERROR(EAGAIN) || ret == AVERROR_EOF)
break;
else if (ret < 0) {
LOGD("Error during decoding\n");
return -1;
}
int data_size = av_get_bytes_per_sample(input_codec_context->sample_fmt);
if (data_size < 0) {
/* This should not occur, checking just for paranoia */
LOGD("Failed to calculate data size\n");
return -1;
}
memcpy(outbuf+index, output_packet.data, output_packet.size);
index+=output_packet.size;
}
*outlen = index;
return 0;
}
int close_aac_codec(void){
if (input_codec_frame != NULL){
av_frame_free(&input_codec_frame);
input_codec_frame = NULL;
}
if (input_codec_context != NULL){
avcodec_free_context(&input_codec_context);
}
return 0;
}
本文为呱牛笔记原创文章,转载无需和我联系,但请注明来自呱牛笔记 ,it3q.com
- 上一篇： 总结2022年
- 下一篇： 音频广播播放功能实现逻辑分享
请先登录后发表评论
- 最新评论
- 总共0条评论
