SpleeterRT with Python

[Matheusadler]

First of all, congrats for your work!

Would it be possible to create a process similar to the VST Plugin you created, but using only ML with Python? I'm experimenting with a use case to create karaoke videos and one of the steps is separating the vocals from the audio tracks, in this case, only separating into 2 stems is needed (Vocals and Accompaniment). I'm trying to optimize the pre-trained models (Spleeter, for example) for the lowest possible latency, and I found your approach interesting. However, I don't have much knowledge in C and plugin creation.

Thank you for your attention :)