I've been working on a karaoke game idea in Python for a bit now and I stumbled upon a small problem:
The scoring system.
As of now, I am able to generate an isolated vocal track and also generate the pitches for this track. So the next step, naturally, would be to figure out how to get the pitches from a live audio input stream.
But given the live audio stream, I would have to be able to compare these pitches live and return that result to the user in some sort of scoring system. Furthermore, the library that is used for determining the pitches in the audio takes a bit of time (using CREPE). It is fine for processing to take a long time when generating a reference track but in practice, there needs to be a faster way.
Well, for now I will tinker with PyAudio for a bit and see what I can do.
Top comments (0)