Multiple transcribe model

We will add the ability to use diffrent models for transcribing audio files