Stem Splitter
Separate any song into individual stems using AI. Runs on your device — files never uploaded.
Drop audio file here or click to browse
MP3, WAV, FLAC, OGG, M4A · Up to 15 minutes
Private — processed on your device, never uploaded
First use downloads a 67 MB AI model (cached for future visits).
Preparing...
Processing locally
How to Split Stems
- 1 Upload a song (MP3, WAV, FLAC, OGG, M4A)
- 2 AI separates vocals from the instrumental
- 3 Solo, mute, or download each stem as WAV
Frequently Asked Questions
How many stems does it separate?
Four stems: vocals, drums, bass, and other instruments. Each is downloadable as a lossless WAV file. You can solo, mute, and adjust the volume of each stem independently.
What is the difference between stem splitting and vocal removal?
Vocal removal gives you two tracks — vocals and everything else. Stem splitting gives you four separate tracks: vocals, drums, bass, and other instruments. Use stem splitting when you need granular control over individual elements.
Is my audio uploaded to a server?
No. All processing runs directly on your device using your own GPU or CPU. Your files never leave your machine — not even temporarily.
How long does stem separation take?
Depends on your device. Desktop with GPU: 5-15 seconds for a typical song. CPU-only: 30-90 seconds. A 67 MB AI model downloads on first use (cached for future visits).
Can I use the stems in my DAW?
Yes. All stems export as uncompressed WAV files at the original sample rate. Import them directly into Ableton, FL Studio, Logic, Pro Tools, or any DAW.
