stem-separation

To isolate musical stems from mixed audio, you can use AI stem separation to split tracks into individual components. This Skill supports extracting vocals, drums, bass, and other instruments to enable remixing, mastering, and post-production workflows.

For audio stem separation, Demucs v4 hybrid transformer provides general multi-instrument extraction, while BSRNN offers improved vocal isolation. This Skill supports both backends to handle different music production and research workflow requirements.

Yes, you can extract guitar and piano stems using the 6-stem separation configuration. Beyond the standard 4-stem output of vocals, drums, bass, and other, this Skill supports a 6-stem mode to isolate guitar and piano for detailed musical analysis.

You can batch process a large music collection for stem separation using the Python API or CLI. This Skill supports GPU acceleration and batch processing to efficiently isolate stems across multiple audio files for large-scale music production workflows.

A GPU is not strictly required to process audio tracks for stem separation, but it is supported. This Skill utilizes GPU acceleration to significantly improve batch processing performance and speed for large music collections and complex AI models.

The best way to integrate stem separation into a Python workflow is using the programmatic API. This Skill provides a Python interface to process single tracks or batches, allowing you to automate stem extraction directly within music production and research scripts.