What's currently a good way to generate transcripts from long video/audio files? I tried using the speechrecognition python library and crashed my computer 😂