For one part of the Android app I am developing, I am wanting to perform some analysis (probably only an STFT) on a previously recorded speech signal. So far I used the MediaRecorder with THREE_GPP as the OutputFormat for recording. I was then planning to use the librosa library to perform the STFT. (I am using Chaquopy to include Python in my code of the app).
However, I keep getting a NoBackendError when loading the audio signal with librosa, which according to other posts is most likely linked to librosa not being able to load the used audio format. So which OutputFormat or AudioEncoder can I use that will be supported by librosa? Or do I have to switch to AudioRecord instead of the MediaRecorder to record in a compatible format?
