AAC 格式分析

一直在做一个语音项目，到了测试阶段，近来不是很忙，想把之前做的内容整理一下。

 

关于AAC音频格式基本情况，可参考维基百科http://en.wikipedia.org/wiki/Advanced_Audio_Coding

 

AAC音频格式分析

AAC音频格式有ADIF和ADTS：

ADIF：Audio Data Interchange Format 音频数据交换格式。这种格式的特征是可以确定的找到这个音频数据的开始，不需进行在音频数据流中间开始的解码，即它的解码必须在明确定义的开始处进行。故这种格式常用在磁盘文件中。

ADTS：Audio Data Transport Stream 音频数据传输流。这种格式的特征是它是一个有同步字的比特流，解码可以在这个流中任何位置开始。它的特征类似于mp3数据流格式。

简单说，ADTS可以在任意帧解码，也就是说它每一帧都有头信息。ADIF只有一个统一的头，所以必须得到所有的数据后解码。且这两种的header的格式也是不同的，目前一般编码后的和抽取出的都是ADTS格式的音频流。

语音系统对实时性要求较高，基本是这样一个流程，采集音频数据，本地编码，数据上传，服务器处理，数据下发，本地解码

ADTS是帧序列，本身具备流特征，在音频流的传输与处理方面更加合适。

 

 

ADTS帧结构：

 

header

body

ADTS帧首部结构：

序号	域	长度（bits）	说明
1	Syncword	12	all bits must be 1
2	MPEG version	1	0 for MPEG-4, 1 for MPEG-2
3	Layer	2	always 0
4	Protection Absent	1	et to 1 if there is no CRC and 0 if there is CRC
5	Profile	2	the MPEG-4 Audio Object Type minus 1
6	MPEG-4 Sampling Frequency Index	4	MPEG-4 Sampling Frequency Index (15 is forbidden)
7	Private Stream	1	set to 0 when encoding, ignore when decoding
8	MPEG-4 Channel Configuration	3	MPEG-4 Channel Configuration (in the case of 0, the channel configuration is sent via an inband PCE)
9	Originality	1	set to 0 when encoding, ignore when decoding
10	Home	1	set to 0 when encoding, ignore when decoding
11	Copyrighted Stream	1	set to 0 when encoding, ignore when decoding
12	Copyrighted Start	1	set to 0 when encoding, ignore when decoding
13	Frame Length	13	this value must include 7 or 9 bytes of header length: FrameLength = (ProtectionAbsent == 1 ? 7 : 9) + size(AACFrame)
14	Buffer Fullness	11	buffer fullness
15	Number of AAC Frames	2	number of AAC frames (RDBs) in ADTS frame minus 1, for maximum compatibility always use 1 AAC frame per ADTS frame
16	CRC	16	CRC if protection absent is 0

 

 

 

AAC解码

在解码方面，使用了开源的FAAD，http://www.audiocoding.com/faad2.html

sdk解压缩后，docs目录有详细的api说明文档，主要用到的有以下几个：

NeAACDecHandle NEAACAPI NeAACDecOpen(void);

创建解码环境并返回一个句柄

void NEAACAPI NeAACDecClose(NeAACDecHandle hDecoder);

关闭解码环境

NeAACDecConfigurationPtr NEAACAPI NeAACDecGetCurrentConfiguration(NeAACDecHandle hDecoder);

获取当前解码器库的配置

unsigned char NEAACAPI NeAACDecSetConfiguration(NeAACDecHandle hDecoder, NeAACDecConfigurationPtr config);

为解码器库设置一个配置结构

long NEAACAPI NeAACDecInit(NeAACDecHandle hDecoder, unsigned char *buffer, unsigned long buffer_size, unsigned long *samplerate, unsigned char *channels);

初始化解码器库

void* NEAACAPI NeAACDecDecode(NeAACDecHandle hDecoder, NeAACDecFrameInfo *hInfo, unsigned char *buffer, unsigned long buffer_size);

解码AAC数据

 

 

对以上api做了简单封装，写了一个解码类，涵盖了FAAD库的基本用法，感兴趣的朋友可以看看

MyAACDecoder.h：

/**
 *
 * filename: MyAACDecoder.h
 * summary: convert aac to wave
 * author: caosiyang
 * email: csy3228@gmail.com
 *
 */
#ifndef __MYAACDECODER_H__
#define __MYAACDECODER_H__


#include "Buffer.h"
#include "mytools.h"
#include "WaveFormat.h"
#include "faad.h"
#include <iostream>
using namespace std;


class MyAACDecoder {
public:
    MyAACDecoder();
    ~MyAACDecoder();

    int32_t Decode(char *aacbuf, uint32_t aacbuflen);

    const char* WavBodyData() const {
        return _mybuffer.Data();
    }

    uint32_t WavBodyLength() const {
        return _mybuffer.Length();
    }

    const char* WavHeaderData() const {
        return _wave_format.getHeaderData();

    }

    uint32_t WavHeaderLength() const {
        return _wave_format.getHeaderLength();
    }

private:
    MyAACDecoder(const MyAACDecoder &dec);
    MyAACDecoder& operator=(const MyAACDecoder &rhs);

    //init AAC decoder
    int32_t _init_aac_decoder(char *aacbuf, int32_t aacbuflen);

    //destroy aac decoder
    void _destroy_aac_decoder();

    //parse AAC ADTS header, get frame length
    uint32_t _get_frame_length(const char *aac_header) const;

    //AAC decoder properties
    NeAACDecHandle _handle;
    unsigned long _samplerate;
    unsigned char _channel;

    Buffer _mybuffer;
    WaveFormat _wave_format;
};


#endif /*__MYAACDECODER_H__*/

 

MyAACDecoder.cpp：

#include "MyAACDecoder.h"


MyAACDecoder::MyAACDecoder(): _handle(NULL), _samplerate(44100), _channel(2), _mybuffer(4096, 4096) {
}


MyAACDecoder::~MyAACDecoder() {
    _destroy_aac_decoder();
}


int32_t MyAACDecoder::Decode(char *aacbuf, uint32_t aacbuflen) {
    int32_t res = 0;
    if (!_handle) {
        if (_init_aac_decoder(aacbuf, aacbuflen) != 0) {
            ERR1(":::: init aac decoder failed ::::");
            return -1;
        }
    }

    //clean _mybuffer
    _mybuffer.Clean();

    uint32_t donelen = 0;
    uint32_t wav_data_len = 0;
    while (donelen < aacbuflen) {
        uint32_t framelen = _get_frame_length(aacbuf + donelen);

        if (donelen + framelen > aacbuflen) {
            break;
        }

        //decode
        NeAACDecFrameInfo info;
        void *buf = NeAACDecDecode(_handle, &info, (unsigned char*)aacbuf + donelen, framelen);
        if (buf && info.error == 0) {
            if (info.samplerate == 44100) {
                //44100Hz
                //src: 2048 samples, 4096 bytes
                //dst: 2048 samples, 4096 bytes
                uint32_t tmplen = info.samples * 16 / 8;
                _mybuffer.Fill((const char*)buf, tmplen);
                wav_data_len += tmplen;
            } else if (info.samplerate == 22050) {
                //22050Hz
                //src: 1024 samples, 2048 bytes
                //dst: 2048 samples, 4096 bytes
                short *ori = (short*)buf;
                short tmpbuf[info.samples * 2];
                uint32_t tmplen = info.samples * 16 / 8 * 2;
                for (int32_t i = 0, j = 0; i < info.samples; i += 2) {
                    tmpbuf[j++] = ori[i];
                    tmpbuf[j++] = ori[i + 1];
                    tmpbuf[j++] = ori[i];
                    tmpbuf[j++] = ori[i + 1];
                }
                _mybuffer.Fill((const char*)tmpbuf, tmplen);
                wav_data_len += tmplen;
            }
        } else {
            ERR1("NeAACDecDecode() failed");
        }

        donelen += framelen;
    }

    //generate Wave header
    _wave_format.setSampleRate(_samplerate);
    _wave_format.setChannel(_channel);
    _wave_format.setSampleBit(16);
    _wave_format.setBandWidth(_samplerate * 16 * _channel / 8);
    _wave_format.setDataLength(wav_data_len);
    _wave_format.setTotalLength(wav_data_len + 44);
    _wave_format.GenerateHeader();

    return 0;
}


uint32_t MyAACDecoder::_get_frame_length(const char *aac_header) const {
    uint32_t len = *(uint32_t *)(aac_header + 3);
    len = ntohl(len); //Little Endian
    len = len << 6;
    len = len >> 19;
    return len;
}


int32_t MyAACDecoder::_init_aac_decoder(char* aacbuf, int32_t aacbuflen) {
    unsigned long cap = NeAACDecGetCapabilities();
    _handle = NeAACDecOpen();
    if (!_handle) {
        ERR1("NeAACDecOpen() failed");
        _destroy_aac_decoder();
        return -1;
    }

    NeAACDecConfigurationPtr conf = NeAACDecGetCurrentConfiguration(_handle);
    if (!conf) {
        ERR1("NeAACDecGetCurrentConfiguration() failed");
        _destroy_aac_decoder();
        return -1;
    }
    NeAACDecSetConfiguration(_handle, conf);

    long res = NeAACDecInit(_handle, (unsigned char *)aacbuf, aacbuflen, &_samplerate, &_channel);
    if (res < 0) {
        ERR1("NeAACDecInit() failed");
        _destroy_aac_decoder();
        return -1;
    }
    //fprintf(stdout, "SampleRate = %d\n", _samplerate);
    //fprintf(stdout, "Channel    = %d\n", _channel);
    //fprintf(stdout, ":::: init aac decoder done ::::\n");

    return 0;
}


void MyAACDecoder::_destroy_aac_decoder() {
    if (_handle) {
        NeAACDecClose(_handle);
        _handle = NULL;
    }
}

AAC 格式分析

标签：

原文地址：http://www.cnblogs.com/xiaofeifei/p/4493152.html