ExtAudioFileWrite创建静默m4a

Question

我正在制作一个应用程序，需要一个可变数量的音频文件合并到一个。为了实现这一点，我正在使用多声道混音单元.混频器成功地混合了音轨，如果我将AUGraph中的I/O单元设置为RemoteIO，它就能成功地播放声音。

但是，我想将新声音保存到一个文件中；我认为最好的方法是使用泛型输出而不是RemoteIO。该文件已成功写入，并且是正确的长度(以秒为单位)，但在播放时是无声的。下面是我如何创建AUGraph的方法(为了简洁起见，我将尽可能删除)

- (void)createGraph
{
    OSStatus result = NewAUGraph(&graph);

    AudioComponentDescription mixerDescription;
    mixerDescription.componentType = kAudioUnitType_Mixer;
    mixerDescription.componentSubType = kAudioUnitSubType_MultiChannelMixer;
    mixerDescription.componentManufacturer = kAudioUnitManufacturer_Apple;
    mixerDescription.componentFlags = 0;
    mixerDescription.componentFlagsMask = 0;

    AudioComponentDescription remoteIODescription;
    remoteIODescription.componentType = kAudioUnitType_Output;
    remoteIODescription.componentSubType = kAudioUnitSubType_GenericOutput;
    remoteIODescription.componentManufacturer = kAudioUnitManufacturer_Apple;
    remoteIODescription.componentFlags = 0;
    remoteIODescription.componentFlagsMask = 0;

    AUNode iONode, mixerNode;

    result = AUGraphAddNode(graph, &remoteIODescription, &iONode);
    result = AUGraphAddNode(graph, &mixerDescription, &mixerNode);
    result = AUGraphOpen(graph);
    result = AUGraphNodeInfo(graph, mixerNode, NULL, &mixerUnit);
    result = AUGraphNodeInfo(graph, iONode, NULL, &iOUnit);
    result = AUGraphConnectNodeInput(graph, mixerNode, 0, iONode, 0);

    UInt32 busCount = (UInt32)fileCount;

    result = AudioUnitSetProperty(mixerUnit, kAudioUnitProperty_ElementCount, kAudioUnitScope_Input, 0, &busCount, sizeof(busCount));

    UInt32 maximumFramesPerSlice = 4096;

    result = AudioUnitSetProperty(mixerUnit, kAudioUnitProperty_MaximumFramesPerSlice, kAudioUnitScope_Global, 0, &maximumFramesPerSlice, sizeof(maximumFramesPerSlice));

    for (UInt16 busNumber = 0; busNumber < busCount; busNumber++)
    {
        AURenderCallbackStruct renderCallback;
        renderCallback.inputProc = &inputRenderCallback;
        renderCallback.inputProcRefCon = (__bridge void *)self;

        result = AUGraphSetNodeInputCallback(graph, mixerNode, busNumber, &renderCallback);

        //sets certain parameters for the mixer; I don't believe this is the cause of the problem
        AudioUnitSetParameter(mixerUnit, kMultiChannelMixerParam_Pan, kAudioUnitScope_Input, busNumber, fileSettings[busNumber].pan, 0);
        AudioUnitSetParameter(mixerUnit, kMultiChannelMixerParam_Volume, kAudioUnitScope_Input, busNumber, fileSettings[busNumber].volume, 0);
        AudioUnitSetParameter(mixerUnit, kMultiChannelMixerParam_Enable, kAudioUnitScope_Input, busNumber, fileSettings[busNumber].enabled, 0);

        if (soundStructs[busNumber].isStereo == YES)
        {
            result = AudioUnitSetProperty(mixerUnit, kAudioUnitProperty_StreamFormat, kAudioUnitScope_Input, busNumber, &stereoDescription, sizeof(stereoDescription));
        }
        else
        {
            result = AudioUnitSetProperty(mixerUnit, kAudioUnitProperty_StreamFormat, kAudioUnitScope_Input, busNumber, &monoDescription, sizeof(monoDescription));
        }
    }

    Float64 sampleRate = SAMPLE_RATE; // 44100.0
    result = AudioUnitSetProperty(mixerUnit, kAudioUnitProperty_SampleRate, kAudioUnitScope_Output, 0, &sampleRate, sizeof(sampleRate));

    CFURLRef url = (__bridge CFURLRef)[NSURL fileURLWithPath:ofn];

    AudioStreamBasicDescription destinationFormat;
    memset(&destinationFormat, 0, sizeof(destinationFormat));
    destinationFormat.mChannelsPerFrame = 2;
    destinationFormat.mFormatID = kAudioFormatMPEG4AAC;
    destinationFormat.mFormatFlags = kMPEG4Object_AAC_Main;
    destinationFormat.mSampleRate = sampleRate;

    UInt32 size = sizeof(destinationFormat);
    result = AudioFormatGetProperty(kAudioFormatProperty_FormatInfo, 0, NULL, &size, &destinationFormat);

    result = ExtAudioFileCreateWithURL(url, kAudioFileM4AType, &destinationFormat, NULL, kAudioFileFlags_EraseFile, &outputFile);

    AudioStreamBasicDescription clientFormat;
    memset(&clientFormat, 0, sizeof(clientFormat));

    size = sizeof(clientFormat);
    result = AudioUnitGetProperty(iOUnit, kAudioUnitProperty_StreamFormat, kAudioUnitScope_Output, 0, & clientFormat, &size);

    UInt32 codec = kAppleHardwareAudioCodecManufacturer;
    ExtAudioFileSetProperty(outputFile, kExtAudioFileProperty_CodecManufacturer, sizeof(codec), &codec);

    ExtAudioFileSetProperty(outputFile, kExtAudioFileProperty_ClientDataFormat, sizeof(clientFormat), &clientFormat);

    result = AUGraphInitialize(graph);
}

我就是这样保存文件的

- (void)startGraph
{
    AudioUnitRenderActionFlags flags = 0;
    AudioTimeStamp inTimeStamp;
    memset(&inTimeStamp, 0, sizeof(AudioTimeStamp));
    inTimeStamp.mFlags = kAudioTimeStampSampleTimeValid;
    UInt32 busNumber = 0;
    UInt32 numberFrames = 1024;
    inTimeStamp.mSampleTime = 0;
    int channelCount = 2;
    SInt64 totFrms = 0;

    for (int i = 0; i < fileCount; i++) //gets the length for the longest recording in the new track
    {
        SInt64 len = soundStructs[i].totalFrames;
        if (len > totFrms && fileSettings[i].enabled == YES)
            totFrms = len;
    }
    while (totFrms > 0)
    {
        if (totFrms < numberFrames)
            numberFrames = (UInt32)totFrms;
        else
            totFrms -= numberFrames;
        AudioBufferList *bufferList = (AudioBufferList*)malloc(sizeof(AudioBufferList)+sizeof(AudioBuffer)*(channelCount-1));
        bufferList->mNumberBuffers = channelCount;
        for (int j=0; j<channelCount; j++)
        {
            AudioBuffer buffer = {0};
            buffer.mNumberChannels = 1;
            buffer.mDataByteSize = numberFrames*sizeof(SInt32);
            buffer.mData = calloc(numberFrames, sizeof(SInt32));

            bufferList->mBuffers[j] = buffer;
        }
        AudioUnitRender(iOUnit, &flags, &inTimeStamp, busNumber, numberFrames, bufferList);

        OSStatus res = ExtAudioFileWrite(outputFile, numberFrames, bufferList);
        NSAssert(res == noErr, @"Res != noerr");
    }

    ExtAudioFileDispose(outputFile);
}

因为这个问题已经很长了，所以我不会添加混频器用于输入的回调函数，也不会添加用于将文件加载到内存中的方法，因为RemoteIO工作的事实使我相信这些函数没有什么问题。那么，为什么输出文件是由我的图形无声创建的呢？

Answer 1

在调用AudioUnitRender之后，需要为下一次迭代增加时间戳的示例时间。

AudioUnitRender(iOUnit, &flags, &inTimeStamp, busNumber, numberFrames, bufferList);
inTimeStamp.mSampleTime += numberFrames;