Open Issues Need Help
View All on GitHubAI Summary: Implement chunked audio processing to improve the memory efficiency of the YouTube audio stem separation application. This involves splitting long audio files into smaller segments using FFmpeg, processing each segment with Spleeter, and then recombining the results using FFmpeg. The goal is to reduce memory usage and increase stability when handling long audio files, while accepting a potential increase in processing time and the need to mitigate potential audio artifacts.
Web API for extracting audio stems from YouTube using Spleeter. Ideal for quick backing track generation. Powered by FastAPI, Celery, and pytubefix.