DE

Demucs

State-of-the-art music source separation

Self-Hosted AI ★ 3k stars Medium setup MIT

Demucs is a deep learning model from Meta Research that separates music into individual stems such as vocals, drums, bass, and other instruments. It runs locally for self-hosted audio processing pipelines.

Key features

  • Multi-stem separation
  • High separation quality
  • CPU and GPU support
  • Command-line interface

Strengths

  • Released under the MIT license
  • Active community (3k GitHub stars)
  • Written in Python

Demucs replaces

Compare Demucs

10 head-to-head comparisons.

Similar self-hosted ai apps