🎚️ AI Stem Splitter
Split music into vocals, drums, bass and other stems.
Choose a file
Choose a music or video file
Separate music into four tracks with AI Stem Splitter
AI Stem Splitter estimates four components of a finished song: vocals, drums, bass and other. Each component can be played separately and downloaded as a WAV file. This helps you listen to parts of an arrangement independently.
What can you use it for?
For your own song, isolate the bass line or drum rhythm to study how it works. You can also import the downloaded files into an audio editor for uses covered by your permissions.
How to use it
- Choose a music recording.
- Start AI separation.
- Listen to the required tracks in the four-part result.
- Download vocals, drums, bass or other as separate WAV files.
Tips for better results
Begin with a short excerpt containing several instruments. The large model and multiple output tracks can use substantial memory; WebGPU support can speed up processing.
Limits and checking the output
The other track groups several kinds of instruments together. This four-category output does not automatically isolate every guitar, piano or synthesiser. Bleed and separation errors can occur.
Frequently asked questions
Will I get separate guitar and piano tracks?
The current tool provides four categories. Guitar and piano will generally be grouped in other.
Are these the original studio stems?
No. The AI estimates components from the final mix, so the results can differ from the original tracks.
Processing and privacy
Your selected file is processed in your browser on your own device; it is not sent to an external AI processing service. Loading the required model and runtime for the first use can take time and consume data. Speed depends on recording length, browser support and device resources. On mobile, try a short excerpt first, keep the tab open and save the result. See the privacy notice for other data processing on the website. Privacy notice.