crsr

há 4 mesesdiscuss

prior version used modal to host the sam audio model (https://github.com/sambarrowclough/clearaudio/pull/2). fal made things simpler & faster

crsr

há 4 mesesdiscuss

alternative open source version here: https://github.com/sambarrowclough/clearaudio

crsr

há 7 mesesdiscuss

I built ClearAudio to make prompt-based audio cleanup practical. Upload audio, describe what you want ("keep the voice, remove the crowd"), and it separates it out. I stripped it down to the essentials with a minimal 1-bit UI—no extra features, just the core separation.

Uses Meta's AudioSeg model (https://github.com/facebookresearch/audioseg) under the hood. Also used this as a chance to learn Modal for GPU inference.

Code is open source: https://github.com/sambarrowclough/clearaudio

Happy to answer questions about the stack or take feedback on the UX!

crsr

há 9 mesesdiscuss

Browserless have browsers as a service, and a dedicated pdf endpoint[0] you can call. Had really good experience with this.

[0] http://docs.browserless.io/rest-apis/pdf-api

Submissions

Show HN: ClearAudio – Isolate sounds from audio using text prompts

comments