Audio recorded using Audio Queue Services to data

Question

I want to transmit voice from one iPhone to another. I have established connection between two iPhones using TCP and I have managed to record voice on the iPhone and play it using Audio Queue Services. I have also managed to send data between the two iPhones. I do this by sending NSData packages.

My next step is to send the audio data to the other iPhone as it is being recorded. I believe I should do this in the AudioInputCallback. My AudioQueueBufferRef is called inBuffer and it seems that I want to convert the inBuffer->mAudioData to NSData and then send the NSData to the other device and then unpack it.

Does anyone know if this would be the way to do it and how I can convert my inBuffer->mAudioData to NSData? Other approaches are also welcome.

This is my callback method in which I believe I should "grab" the data and send it to the other iPhone:

void AudioInputCallback(void *inUserData, AudioQueueRef inAQ, AudioQueueBufferRef inBuffer, const AudioTimeStamp *inStartTime, UInt32 inNumberPacketDescriptions, const AudioStreamPacketDescription *inPacketDescs)
{
    RecordState *recordState = (RecordState *)inUserData;
    if(!recordState->recording)
        return;

    OSStatus status = AudioFileWritePackets(recordState->audioFile,
                                            false,
                                            inBuffer->mAudioDataByteSize,
                                            inPacketDescs,
                                            recordState->currentPacket,
                                            &inNumberPacketDescriptions,
                                            inBuffer->mAudioData);
    if(status == 0)
    {
        recordState->currentPacket += inNumberPacketDescriptions;
    }

    AudioQueueEnqueueBuffer(recordState->queue, inBuffer, 0, NULL);
}

Answer 1

You might want to consider saving the audio data (your example shows the audio sample pointer and the byte count) from the audio callback to another queue or FIFO, then having a separate networking thread create NSData from the audio bytes and sending it.