I want to Capture the audio being played by other apps using AudioPlaybackCapture API and load it into a classifier but I cant load the AudioRecord directly to TensorAudio because it throws unsupported format error. my only chance is to convert AudioRecord to short array or float array and load it into classifier.