多媒体开发（13）：iOS上音频编码成aac

如前面我所说，对于音频的解码，一般你都不用考虑硬解，用软解就足够了，这时可以选择faad或FFmpeg等。但是，如果是音频的编码呢？这可不一样，编码比解码明显耗时，为了快跟低功耗（特别对于低端机器），要优先考虑硬编码（不能再使用fdk-aac或faac之类的软编码），硬编码的优势是可以用硬件芯片集成的功能，高速且低功耗地完成编码任务。

iOS平台，也提供了硬编码的能力，APP开发时只需要调用相应的SDK接口就能达成目标，这个SDK接口就是AudioConverter。

本文介绍iOS平台上，如何调用AudioConverter来完成aac的硬编码。

从名字来看，AudioConverter就是格式转换器，那就对了，这里把pcm格式的数据，转换成aac格式的数据。

AudioConverter在内存中实现转换，并不需要写文件，而ExtAudioFile接口则是对文件的操作，并且内部使用AudioConerter来转换格式，也就是说，你在某种场景下，也可以使用ExtAudioFile接口并接受临时文件的过程。

要独立操作，就要理解细节。具体如何使用AudioConverter呢？基本上，对接口的调用都需要阅读对应的头文件，通过看文档注释来理解怎么调用。

小程这里演示一下，怎么把pcm转换成aac。在演示代码之后，我只做简单的解释，如果你有需要，请耐心阅读代码来理解，并应用到自己的开发场景中。

typedef struct
{
void *source;
UInt32 sourceSize;
UInt32 channelCount;
AudioStreamPacketDescription *packetDescriptions;
}FillComplexInputParam;

// 填写源数据，即pcm数据
OSStatus audioConverterComplexInputDataProc(  AudioConverterRef               inAudioConverter,
UInt32*                         ioNumberDataPackets,
AudioBufferList*                ioData,
AudioStreamPacketDescription**  outDataPacketDescription,
void*                           inUserData)
{
FillComplexInputParam* param = (FillComplexInputParam*)inUserData;
if (param->sourceSize <= 0) {
*ioNumberDataPackets = 0;
return -1;
}
ioData->mBuffers[0].mData = param->source;
ioData->mBuffers[0].mNumberChannels = param->channelCount;
ioData->mBuffers[0].mDataByteSize = param->sourceSize;
*ioNumberDataPackets = 1;
param->sourceSize = 0;
param->source = NULL;
return noErr;
}

typedef struct _tagConvertContext {
AudioConverterRef converter;
int samplerate;
int channels;
}ConvertContext;

// init
// 最终用AudioConverterNewSpecific创建ConvertContext，并设置比特率之类的属性
void* convert_init(int sample_rate, int channel_count)
{
AudioStreamBasicDescription sourceDes;
memset(&sourceDes, 0, sizeof(sourceDes));
sourceDes.mSampleRate = sample_rate;
sourceDes.mFormatID = kAudioFormatLinearPCM;
sourceDes.mFormatFlags = kLinearPCMFormatFlagIsPacked | kLinearPCMFormatFlagIsSignedInteger;
sourceDes.mChannelsPerFrame = channel_count;
sourceDes.mBitsPerChannel = 16;
sourceDes.mBytesPerFrame = sourceDes.mBitsPerChannel/8*sourceDes.mChannelsPerFrame;
sourceDes.mBytesPerPacket = sourceDes.mBytesPerFrame;
sourceDes.mFramesPerPacket = 1;
sourceDes.mReserved = 0;

AudioStreamBasicDescription targetDes;
memset(&targetDes, 0, sizeof(targetDes));
targetDes.mFormatID = kAudioFormatMPEG4AAC;
targetDes.mSampleRate = sample_rate;
targetDes.mChannelsPerFrame = channel_count;
UInt32 size = sizeof(targetDes);
AudioFormatGetProperty(kAudioFormatProperty_FormatInfo, 0, NULL, &size, &targetDes);

AudioClassDescription audioClassDes;
memset(&audioClassDes, 0, sizeof(AudioClassDescription));
AudioFormatGetPropertyInfo(kAudioFormatProperty_Encoders, sizeof(targetDes.mFormatID), &targetDes.mFormatID, &size);
int encoderCount = size / sizeof(AudioClassDescription);
AudioClassDescription descriptions[encoderCount];
AudioFormatGetProperty(kAudioFormatProperty_Encoders, sizeof(targetDes.mFormatID), &targetDes.mFormatID, &size, descriptions);
for (int pos = 0; pos < encoderCount; pos ++) {
if (targetDes.mFormatID == descriptions[pos].mSubType && descriptions[pos].mManufacturer == kAppleSoftwareAudioCodecManufacturer) {
memcpy(&audioClassDes, &descriptions[pos], sizeof(AudioClassDescription));
break;
}
}

ConvertContext *convertContex = malloc(sizeof(ConvertContext));
OSStatus ret = AudioConverterNewSpecific(&sourceDes, &targetDes, 1, &audioClassDes, &convertContex->converter);
if (ret == noErr) {
AudioConverterRef converter = convertContex->converter;

tmp = kAudioConverterQuality_High;
AudioConverterSetProperty(converter, kAudioConverterCodecQuality, sizeof(tmp), &tmp);

UInt32 bitRate = 96000;
UInt32 size = sizeof(bitRate);
ret = AudioConverterSetProperty(converter, kAudioConverterEncodeBitRate, size, &bitRate);
}
else {
free(convertContex);
convertContex = NULL;
}

return convertContex;
}

// converting
void convert(void* convertContext, void* srcdata, int srclen, void** outdata, int* outlen)
{
ConvertContext* convertCxt = (ConvertContext*)convertContext;
if (convertCxt && convertCxt->converter) {
UInt32 theOuputBufSize = srclen;
UInt32 packetSize = 1;
void *outBuffer = malloc(theOuputBufSize);
memset(outBuffer, 0, theOuputBufSize);

AudioStreamPacketDescription *outputPacketDescriptions = NULL;
outputPacketDescriptions = (AudioStreamPacketDescription*)malloc(sizeof(AudioStreamPacketDescription) * packetSize);

FillComplexInputParam userParam;
userParam.source = srcdata;
userParam.sourceSize = srclen;
userParam.channelCount = convertCxt->channels;
userParam.packetDescriptions = NULL;

OSStatus ret = noErr;

AudioBufferList* bufferList = malloc(sizeof(AudioBufferList));
AudioBufferList outputBuffers = *bufferList;
outputBuffers.mNumberBuffers = 1;
outputBuffers.mBuffers[0].mNumberChannels = convertCxt->channels;
outputBuffers.mBuffers[0].mData = outBuffer;
outputBuffers.mBuffers[0].mDataByteSize = theOuputBufSize;
ret = AudioConverterFillComplexBuffer(convertCxt->converter, audioConverterComplexInputDataProc, &userParam, &packetSize, &outputBuffers, outputPacketDescriptions);
if (ret == noErr) {
if (outputBuffers.mBuffers[0].mDataByteSize > 0) {

NSData* rawAAC = [NSData dataWithBytes:outputBuffers.mBuffers[0].mData length:outputBuffers.mBuffers[0].mDataByteSize];
*outdata = malloc([rawAAC length]);
memcpy(*outdata, [rawAAC bytes], [rawAAC length]);
*outlen = (int)[rawAAC length];
// 测试转换出来的aac数据，保存成adts-aac文件
#if 1
int headerLength = 0;
char* packetHeader = newAdtsDataForPacketLength((int)[rawAAC length], convertCxt->samplerate, convertCxt->channels, &headerLength);
NSData* adtsPacketHeader = [NSData dataWithBytes:packetHeader length:headerLength];
free(packetHeader);
NSMutableData* fullData = [NSMutableData dataWithData:adtsPacketHeader];
[fullData appendData:rawAAC];

NSFileManager *fileMgr = [NSFileManager defaultManager];
NSString *filepath = [NSHomeDirectory() stringByAppendingFormat:@"/Documents/test%p.aac", convertCxt->converter];
NSFileHandle *file = nil;
if (![fileMgr fileExistsAtPath:filepath]) {
[fileMgr createFileAtPath:filepath contents:nil attributes:nil];
}
file = [NSFileHandle fileHandleForWritingAtPath:filepath];
[file seekToEndOfFile];
[file writeData:fullData];
[file closeFile];
#endif
}
}

free(outBuffer);
if (outputPacketDescriptions) {
free(outputPacketDescriptions);
}
}
}

// uninit
// ...

int freqIdxForAdtsHeader(int samplerate)
{
/**
0: 96000 Hz
1: 88200 Hz
2: 64000 Hz
3: 48000 Hz
4: 44100 Hz
5: 32000 Hz
6: 24000 Hz
7: 22050 Hz
8: 16000 Hz
9: 12000 Hz
10: 11025 Hz
11: 8000 Hz
12: 7350 Hz
13: Reserved
14: Reserved
15: frequency is written explictly
*/
int idx = 4;
if (samplerate >= 7350 && samplerate < 8000) {
idx = 12;
}
else if (samplerate >= 8000 && samplerate < 11025) {
idx = 11;
}
else if (samplerate >= 11025 && samplerate < 12000) {
idx = 10;
}
else if (samplerate >= 12000 && samplerate < 16000) {
idx = 9;
}
else if (samplerate >= 16000 && samplerate < 22050) {
idx = 8;
}
else if (samplerate >= 22050 && samplerate < 24000) {
idx = 7;
}
else if (samplerate >= 24000 && samplerate < 32000) {
idx = 6;
}
else if (samplerate >= 32000 && samplerate < 44100) {
idx = 5;
}
else if (samplerate >= 44100 && samplerate < 48000) {
idx = 4;
}
else if (samplerate >= 48000 && samplerate < 64000) {
idx = 3;
}
else if (samplerate >= 64000 && samplerate < 88200) {
idx = 2;
}
else if (samplerate >= 88200 && samplerate < 96000) {
idx = 1;
}
else if (samplerate >= 96000) {
idx = 0;
}

return idx;
}

int channelIdxForAdtsHeader(int channelCount)
{
/**
0: Defined in AOT Specifc Config
1: 1 channel: front-center
2: 2 channels: front-left, front-right
3: 3 channels: front-center, front-left, front-right
4: 4 channels: front-center, front-left, front-right, back-center
5: 5 channels: front-center, front-left, front-right, back-left, back-right
6: 6 channels: front-center, front-left, front-right, back-left, back-right, LFE-channel
7: 8 channels: front-center, front-left, front-right, side-left, side-right, back-left, back-right, LFE-channel
8-15: Reserved
*/
int ret = 2;
if (channelCount == 1) {
ret = 1;
}
else if (channelCount == 2) {
ret = 2;
}

return ret;
}

/**
*  Add ADTS header at the beginning of each and every AAC packet.
*  This is needed as MediaCodec encoder generates a packet of raw
*  AAC data.
*
*  Note the packetLen must count in the ADTS header itself.
*  See: http://wiki.multimedia.cx/index.php?title=ADTS
*  Also: http://wiki.multimedia.cx/index.php?title=MPEG-4_Audio#Channel_Configurations
**/
char* newAdtsDataForPacketLength(int packetLength, int samplerate, int channelCount, int* ioHeaderLen) {
int adtsLength = 7;
char *packet = malloc(sizeof(char) * adtsLength);
// Variables Recycled by addADTStoPacket
int profile = 2;  //AAC LC
//39=MediaCodecInfo.CodecProfileLevel.AACObjectELD;
int freqIdx = freqIdxForAdtsHeader(samplerate);
int chanCfg = channelIdxForAdtsHeader(channelCount);  //MPEG-4 Audio Channel Configuration.
NSUInteger fullLength = adtsLength + packetLength;
// fill in ADTS data
packet[0] = (char)0xFF;
// 11111111  = syncword
packet[1] = (char)0xF9;
// 1111 1 00 1  = syncword MPEG-2 Layer CRC
packet[2] = (char)(((profile-1)<<6) + (freqIdx<<2) +(chanCfg>>2));
packet[3] = (char)(((chanCfg&3)<<6) + (fullLength>>11));
packet[4] = (char)((fullLength&0x7FF) >> 3);
packet[5] = (char)(((fullLength&7)<<5) + 0x1F);
packet[6] = (char)0xFC;
*ioHeaderLen = adtsLength;
return packet;
}

以上代码，有两个函数比较重要，一个是初始化函数，这个函数创建了AudioConverterRef，另一个是转换函数，这个函数应该被反复调用，对不同的pcm数据进行转换。

另外，示例中，把pcm转换出来的aac数据，进行了保存，保存出来的文件可以用于播放。注意，AudioConverter转换出来的都是音频裸数据，至于组合成adts-aac，还是封装成苹果的m4a文件，由你的程序决定。

这里解释一下，adts-aac是aac数据的一种表示方式，也就是在每帧aac裸数据前面，增加一个帧信息（包括每帧的长度、采样率、声道数等），加上帧信息后，每帧aac可以单独播放。而且，adts-aac没有特定的文件头以及文件结构等。adts是Audio Data Transport Stream的缩写。

当然，你也可以把转换出来的aac数据，封装成m4a格式，这种封装格式，先是文件头，然后是box的组合（包括音频数据mdat等），可参考mp4封装格式。

至此，iOS平台把pcm转换成aac数据的实现就介绍完毕了。

总结一下，本文介绍了如何使用iOS平台提供的AudioConverter接口，把pcm格式的数据转换成aac格式。文章也介绍了怎么保存成adts-aac文件，你可以通过这个办法检验转换出来的aac数据是否正确。