AudioFormat conversion support

Currently it is required to pass a format when recording/playing audio, which should describe the format of the audio collected from the audio data flow. However, some platforms do not offer a lot of flexibility in what audio is supported or not. Some platforms do offer automatic conversion, but others don't.

So perhaps we'd need to have a common post processor that can convert the returned audio data into the requested format. That way we don't have to rely on platform specific audio format conversions.