I’ve been having some odd behaviors with my audio engine since the iPhone 6s and 6s+ have come out and I was hoping someone could help explain these points of confusion for me. I’m quite sure it’s because these new models have a microphone with a set sample rate of 48kHz, especially since plugging in the apple earpods mic fixes the problem, but I still am having the following problems:
First, I am having some unintentional sample rate conversions happening somewhere in my code. I have an AVAudioSession that maintains a 48kHz hardwareSampleRate, which my engine would happily accept, and yet the CMSampleBuffers that I am retrieving through my AVCaptureAudioDataOutputSampleBufferDelegate’s captureOutput function are coming in at 44.1kHz with 940 or 941 samples per buffer, as if there was some sample rate conversion happening before I captured the buffer. Second, what is determining the amount of frames that an Audio Unit’s render callback is allocating for the incoming buffer? I am getting very inconsistent results with this and can not understand how it is being determined. It is represented by the inNumberFrames parameter in an AURenderCallback function. For example, when I start recording it will give me a buffer for 941 samples then immediately increase to giving me a buffer for 1010, 1012, or even 1024 samples. This is causing all sorts of havoc with my render cycles. _______________________________________________ Do not post admin requests to the list. They will be ignored. Coreaudio-api mailing list ([email protected]) Help/Unsubscribe/Update your Subscription: https://lists.apple.com/mailman/options/coreaudio-api/archive%40mail-archive.com This email sent to [email protected]
