after i have put together all this code , how would i get the actual audio data ? where exactly does it saved ?

and when all this code is done ? i have to put it all into 1 method and then call it ? or should i only call -start method ?

what operation should be taken to get the real time data ?

i have spent days to understand it but i couldnt . how exactly i

audioFormat.mSampleRate         = 8000.0;
audioFormat.mFormatID           = kAudioFormatLinearPCM;
audioFormat.mFormatFlags        = kAudioFormatFlagIsSignedInteger | kAudioFormatFlagIsPacked;
audioFormat.mFramesPerPacket    = 1;
audioFormat.mChannelsPerFrame   = 1;
audioFormat.mBitsPerChannel     = 16;
audioFormat.mBytesPerPacket     = 2;
audioFormat.mBytesPerFrame      = 2;
audioFormat.mReserved           = 0;

actually I wanted to play amr file .So I use decoder .It worked,but very noisy and abnormal.

static OSStatus playbackCallback(void *inRefCon, AudioUnitRenderActionFlags *ioActionFlags, const AudioTimeStamp *inTimeStamp, UInt32 inBusNumber, UInt32 inNumberFrames, AudioBufferList *ioData) { //loop through all the buffers that need to be filled for (int i = 0 ; i < ioData->mNumberBuffers; i++){ AudioBuffer buffer = ioData->mBuffers[i]; UInt32 *frameBuffer = buffer.mData; AmrDecode(frameBuffer); } return noErr; }

I really need some help

static OSStatus playbackCallback(void *inRefCon, AudioUnitRenderActionFlags *ioActionFlags, const AudioTimeStamp *inTimeStamp, UInt32 inBusNumber, UInt32 inNumberFrames, AudioBufferList *ioData) { //loop through all the buffers that need to be filled for (int i = 0 ; i < ioData->mNumberBuffers; i++){ AudioBuffer buffer = ioData->mBuffers[i]; UInt32 *frameBuffer = buffer.mData; AmrDecode(frameBuffer); } return noErr; }

I really need some help

In terms of your implementation, firstly, I don’t see any synchronisation going on. That code will run on a separate thread, which means you’ll get all sorts of problems unless you’re careful with how you access the data. You should use a ring buffer or something. If you’re not familiar with multithreading principles, I suggest grabbing a good book or finding a tutorial online, because it’s very, very important to get right. Note, though, that you should avoid holding locks on the high-priority Core Audio thread. A lock-free ring buffer is the best bet. I’ve built one, or you could grab Kurt Revis’ VirtualRingBuffer, which is very, very clever, and I think better than mine.

Also, you’re doing an awful lot of memory allocation there – you should really avoid any kind of allocation in Core Audio’s realtime thread.

If you need further help, I suggest posting on the Core Audio API mailing list.

NSAutoreleasePool *pool = [[NSAutoreleasePool alloc] init];

ShowView *viewController = (ShowView *)inRefCon;

NSData *data = [[NSData alloc] initWithData:viewController->audioData];

if ([data length] > 0) { const int channel = 0; Float32 *buffer = (Float32 *)ioData->mBuffers[channel].mData;

// Generate the samples
for (UInt32 frame = 0; frame < inNumberFrames; frame++) 
{
    NSData *bytedata = [NSData dataWithBytes:[data bytes] + (frame * 4) length:4];
    unsigned char *fileBytes = (unsigned char *)malloc([bytedata length]);
    [bytedata getBytes:fileBytes];
    buffer[frame] = *(Float32*)(fileBytes);
    free(fileBytes);
}

viewController->audioData = nil;

}

[data release];

[pool release];

return noErr;

The audio is playing good. Just there are some constant noise from the background. Any suggestion is appreciated. Thanks!

AudioStreamBasicDescription streamFormat;
streamFormat.mSampleRate = 8000;
streamFormat.mFormatID = kAudioFormatULaw;
streamFormat.mFormatFlags = 0;
streamFormat.mFramesPerPacket = 1;

streamFormat.mBytesPerFrame = 2;
streamFormat.mBytesPerPacket = streamFormat.mBytesPerFrame;
streamFormat.mChannelsPerFrame = 1;
streamFormat.mBitsPerChannel = 16;

I get a constant noise even I just return no error on the playback callback function. Please help. Thanks!

Just want to tell that when you’re finished it is also good to call AudioComponentInstanceDispose(audioUnit);

Firstly, I notice that you’re setting mDataByteSize of the render buffer to inNumberFrames * sizeof(SInt16) * 2 – it may make no difference, but that implies stereo audio (that is, frames * bytes per sample * 2 samples per frame). Either set your mChannelsPerFrame to 2 if stereo audio is what you actually want, or get rid of the * 2.

Also, it couldn’t hurt to make absolutely certain viewController.ioUnit is initialised.

Otherwise, if that doesn’t do it, perhaps it’s worth asking on the Core Audio mailing list, where they’re always very helpful.

OSStatus status = AudioUnitRender(viewController.ioUnit, 
                         ioActionFlags, 
                         inTimeStamp, 
                         1, 
                         inNumberFrames, 
                         &bufferList);

Currently I have a AUFilePlayer->Mixer->iPodSpeed->RemoteIO-Output… this works well, but I want to use files from the iPod Library, or another app.

Thanks