Apple

Hi Alex,

set asbd with kAudioFormatFlagsAudioUnitCanonical and set channels to 2

like this:

size_t bytesPerSample = sizeof (AudioUnitSampleType);

// Fill the application audio format struct's fields to define a linear PCM,

// stereo, noninterleaved stream at the hardware sample rate.

stereoStreamFormat.mFormatID = kAudioFormatLinearPCM;

stereoStreamFormat.mFormatFlags = kAudioFormatFlagsAudioUnitCanonical;

stereoStreamFormat.mBytesPerPacket = bytesPerSample;

stereoStreamFormat.mFramesPerPacket = 1;

stereoStreamFormat.mBytesPerFrame = bytesPerSample;

stereoStreamFormat.mChannelsPerFrame = 2; // 2 indicates stereo

stereoStreamFormat.mBitsPerChannel = 8 * bytesPerSample;

stereoStreamFormat.mSampleRate = graphSampleRate;

then in the callback pull samples from input

AudioUnitRender(rioUnit, ioActionFlags, inTImeStamp, bus1, inNumberFrames, ioData);

you can access left and right channel data like this:

AudioUnitSampleType *left;

AudioUnitSampleType *right;

left = ioData->mBuffers[0].mData;

right = ioData->mBuffers[1].mData;

Check out the apple mixerHost sample code example or: http://zerokidz.com/audiograph

Tom

From: Alex Gross

Sent: Thursday, December 15, 2011 3:16 PM

To: email@hidden

Subject: Stereo recording with remoteIO - how?

I developed an iOS low-latency audio app using remoteIO for 3 years now and I thought I knew my way around AudioUnits. Mono recording works fine, but then I tried for 2 days to get stereo recording working and I had no luck. I tested it with an iPad in the Alesis IO Dock and with GuitarJack on an old iPod Touch. GarageBand for iOS offers channel selection and records stereo just fine. All I'm getting is the left channel.

Here's my question:

In a recording callback

static OSStatus recordingCallback(void *inRefCon, AudioUnitRenderActionFlags *ioActionFlags, const AudioTimeStamp *inTimeStamp, UInt32 inBusNumber, UInt32 inNumberFrames, AudioBufferList *ioData)

{

...

AudioBufferList *inputBufferList = calloc(1, offsetof(AudioBufferList, mBuffers) + (sizeof(AudioBuffer)));

inputBufferList->mNumberBuffers = 1;

inputBufferList->mBuffers[0].mNumberChannels = 2;

inputBufferList->mBuffers[0].mDataByteSize = 2 * sizeof(SInt16) * inNumberFrames;

inputBufferList->mBuffers[0].mData = malloc(inputBufferList->mBuffers[0].mDataByteSize);

AudioUnitRender(audioUnit, ioActionFlags, inTimeStamp, 1, inNumberFrames, inputBufferList); // 1 stands for the input bus

...

}

where during initialization, the format was set to

AudioStreamBasicDescription inputFormat = { 0 };

inputFormat.mSampleRate = 44100;

inputFormat.mFormatID = kAudioFormatLinearPCM;

inputFormat.mFormatFlags = kAudioFormatFlagIsSignedInteger | kAudioFormatFlagIsPacked;

inputFormat.mFramesPerPacket = 1;

inputFormat.mChannelsPerFrame = 2; // of course I detected mono/stereo before and set this to 1 if only mono is available

inputFormat.mBitsPerChannel = 16;

inputFormat.mBytesPerFrame = inputFormat.mChannelsPerFrame * sizeof(SInt16);

inputFormat.mBytesPerPacket = inputFormat.mBytesPerFrame * inputFormat.mFramesPerPacket;

UInt32 inputFormatSize = sizeof(inputFormat);

AudioUnitSetProperty(audioUnit, kAudioUnitProperty_StreamFormat, kAudioUnitScope_Output, 1, &inputFormat, inputFormatSize); // 1 stands for the input bus

How do I retrieve both channels from inputBufferList?

FAIL 1:

Since I didn't specify kAudioFormatFlagIsNonInterleaved, I expected to get 1 buffer with both channels interleaved. Is the size of the buffer (2 * inNumberFrames * sizeof(SInt16)) or (inNumberFrames * sizeof(SInt16)) ?

De-interleaving the data doesn't work:

SInt16 *stereoBuffer = malloc(inputBufferList->mBuffers[0].mDataByteSize);

SInt16 *stereoBufferPtr = stereoBuffer;

SInt16 *leftBufferPtr = inputBufferList->mBuffers[0].mData;

SInt16 *rightBufferPtr = inputBufferList->mBuffers[0].mData;

rightBufferPtr += inNumberFrames;

for(UInt32 i = 0; i < inNumberFrames; ++i)

{

*stereoBufferPtr++ = *leftBufferPtr++;

*stereoBufferPtr++ = *rightBufferPtr++;

}

// ... copying stereoBuffer into my output buffer ...

free(stereoBuffer);

FAIL 2:

I also tried setting kAudioFormatFlagIsNonInterleaved and therefore allocating

inputBufferList->mNumberBuffers = 2;

for(UInt32 i = 0; i < inputBufferList->mNumberBuffers; ++i)

{

inputBufferList->mBuffers[0].mNumberChannels = 1;

inputBufferList->mBuffers[0].mDataByteSize = sizeof(SInt16) * inNumberFrames;

inputBufferList->mBuffers[0].mData = malloc(inputBufferList->mBuffers[0].mDataByteSize);

}

Is it right to assume that I first need to set the kAudioUnitProperty_StreamFormat and then allocate an AudioBufferList, and that both need to be compatible and that they define how AudioUnitRender stores the data into my AudioBufferList?

What am I overlooking or doing wrong? Is there a good example or documentation anywhere on the internet? The aurioTouch example didn't help because it does input and output in the same callback. Thanks in advance for any advice!

_______________________________________________
Do not post admin requests to the list. They will be ignored.
Coreaudio-api mailing list (email@hidden)
Help/Unsubscribe/Update your email sent to email@hidden

References:
	>Stereo recording with remoteIO - how? (From: Alex Gross <email@hidden>)