EZAudio自定义AudioStreamBasicDescription的工作方式不符合我的预期

Question

使用EZAudio，我想尽可能地创建单光audioBufferList。在过去，我实现了每个audioBuffer 46字节，但bufferDuration相对较小。首先，如果我使用下面的AudioStreamBasicDescription作为输入和输出

AudioStreamBasicDescription audioFormat;
     audioFormat.mBitsPerChannel   = 8 * sizeof(AudioUnitSampleType);
     audioFormat.mBytesPerFrame    = sizeof(AudioUnitSampleType);
     audioFormat.mBytesPerPacket   = sizeof(AudioUnitSampleType);
     audioFormat.mChannelsPerFrame = 2;
     audioFormat.mFormatFlags      = kAudioFormatFlagsCanonical | kAudioFormatFlagIsNonInterleaved;
     audioFormat.mFormatID         = kAudioFormatLinearPCM;
     audioFormat.mFramesPerPacket  = 1;
     audioFormat.mSampleRate       = 44100;

并使用TPCircularBuffer作为传输器，然后我在使用mDataByteSize 4096的bufferList中得到两个缓冲区，这绝对是太多了。所以我试着用我以前的ASBD

audioFormat.mSampleRate         = 8000.00;
audioFormat.mFormatID           = kAudioFormatLinearPCM;
audioFormat.mFormatFlags        = kAudioFormatFlagsCanonical | kAudioFormatFlagIsNonInterleaved;
audioFormat.mFramesPerPacket    = 1;
audioFormat.mChannelsPerFrame   = 1;
audioFormat.mBitsPerChannel     = 8;
audioFormat.mBytesPerPacket     = 1;
audioFormat.mBytesPerFrame      = 1;

现在mDataByteSize是128，我只有一个缓冲区，但是TPCircularBuffer不能正确处理这个问题。我想这是因为我只想使用一个通道。所以atm拒绝了TBCB，并尝试将字节编码和解码为NSData，或者只是为了测试直接通过AudioBufferList，但即使是第一个AudioStreamBasicDescription声音也太失真了。

我当前的代码

-(void)initMicrophone{

    AudioStreamBasicDescription audioFormat;
    //*
     audioFormat.mBitsPerChannel   = 8 * sizeof(AudioUnitSampleType);
     audioFormat.mBytesPerFrame    = sizeof(AudioUnitSampleType);
     audioFormat.mBytesPerPacket   = sizeof(AudioUnitSampleType);
     audioFormat.mChannelsPerFrame = 2;
     audioFormat.mFormatFlags      = kAudioFormatFlagsCanonical | kAudioFormatFlagIsNonInterleaved;
     audioFormat.mFormatID         = kAudioFormatLinearPCM;
     audioFormat.mFramesPerPacket  = 1;
     audioFormat.mSampleRate       = 44100;

     /*/
    audioFormat.mSampleRate         = 8000.00;
    audioFormat.mFormatID           = kAudioFormatLinearPCM;
    audioFormat.mFormatFlags        = kAudioFormatFlagsCanonical | kAudioFormatFlagIsNonInterleaved;
    audioFormat.mFramesPerPacket    = 1;
    audioFormat.mChannelsPerFrame   = 1;
    audioFormat.mBitsPerChannel     = 8;
    audioFormat.mBytesPerPacket     = 1;
    audioFormat.mBytesPerFrame      = 1;

    //*/


    _microphone = [EZMicrophone microphoneWithDelegate:self withAudioStreamBasicDescription:audioFormat];

    _output = [EZOutput outputWithDataSource:self withAudioStreamBasicDescription:audioFormat];
    [EZAudio circularBuffer:&_cBuffer withSize:128];
}

-(void)startSending{
    [_microphone startFetchingAudio];
    [_output startPlayback];
}

-(void)stopSending{
    [_microphone stopFetchingAudio];
    [_output stopPlayback];
}

-(void)microphone:(EZMicrophone *)microphone
 hasAudioReceived:(float **)buffer
   withBufferSize:(UInt32)bufferSize
withNumberOfChannels:(UInt32)numberOfChannels{
    dispatch_async(dispatch_get_main_queue(), ^{
    });
}

-(void)microphone:(EZMicrophone *)microphone
    hasBufferList:(AudioBufferList *)bufferList
   withBufferSize:(UInt32)bufferSize
withNumberOfChannels:(UInt32)numberOfChannels{
//*
        abufferlist = bufferList;
    /*/
     audioBufferData = [NSData dataWithBytes:bufferList->mBuffers[0].mData length:bufferList->mBuffers[0].mDataByteSize];
     //*/
 dispatch_async(dispatch_get_main_queue(), ^{
 });
}
-(AudioBufferList*)output:(EZOutput *)output needsBufferListWithFrames:(UInt32)frames withBufferSize:(UInt32 *)bufferSize{
    //*
    return abufferlist;
    /*/
     //    int bSize = 128;
     //    AudioBuffer audioBuffer;
     //    audioBuffer.mNumberChannels = 1;
     //    audioBuffer.mDataByteSize = bSize;
     //    audioBuffer.mData = malloc(bSize);
     ////    [audioBufferData getBytes:audioBuffer.mData length:bSize];
     //    memcpy(audioBuffer.mData, [audioBufferData bytes], bSize);
     //
     //
     //    AudioBufferList *bufferList = [EZAudio audioBufferList];
     //    bufferList->mNumberBuffers = 1;
     //    bufferList->mBuffers[0] = audioBuffer;
     //
     //    return bufferList;
    //*/


}

我知道output:needsBufferListWithFrames:withBufferSize:中bSize的值可能会改变。

我的主要目标是创建光，因为它可以是单声道声音，编码为nsdata，并解码输出。你能告诉我我哪里做错了吗？

Answer 1

我遇到了同样的问题，转到AVAudioRecorder并设置了我需要的参数I kept (EZMicrophone)用于音频可视化，下面是实现这一点的链接：

iOS: Audio Recording File Format