Description
Description
Dear Maintainers,
First of all thank you so much for this library! It is immensely helpful as part of a machine learning research project I am currently working on. It seems like the inner implementations of separators all require the audios to be written to a file. I suppose it could be helpful if we have a method that allows users to input and get outputs in low-level formats like say numpy arrays . This could potentially save audio i/o time and allow for more direct access to the models which could help in large-scaled audio processing projects.
I am a full time research student in computer music and has worked with ML and audio separation models for a while now so I can probably help with this and submit a PR if needed.
Thanks,
Darin