Hi i started audio processing with python and started with this example: https://pytorch.org/audio/main/tutorials/online_asr_tutorial.html
I used streamlink to get an url of a livestream then added it to the src, in the example above. It works but isn't that accurate. I want to try other models like googles. Is there a way to add the audio so it does speech to text in real time like with the example above or do I have to download the stream in chunks then load it in. If I where to load it from chunks i was thinking of using ffmpeg and passing into google's cloud speech to text. Would that be sufficient?