Repo: https://github.com/omeryusufyagci/fast-music-remover
The tool is lightweight, works through Docker, and is licensed under MIT. It supports several use cases, such as improving the audio quality of lectures, field reports, and isolating speech from videos with heavy background noise.
I'm currently working towards adding live stream support and would love to have contributors on board. There are interesting open issues for developers experienced in backend, frontend, Python, C++, and audio processing.
Feedback, suggestions, and contributions are very welcome. Thanks for checking it out!