Universal Stem Splitter: Isolate any instrument from songs

The most advanced stem splitting models for vocals, drums, guitar, bass, and even denoising. Want more? Write the instrument name to use SAM Audio.

SELECT SEPARATION TYPE
Or
Supported links: suno.com, udio.com, producer.ai (songs, playlists, creators)

Next Level Stem Splitting Quality

Extract stems using SAM Audio for flexible custom instrument extraction. Get the best quality using specialized, pretrained models like BS Roformer, MelBand Roformer and MDX23 for state-of-the-art quality.

Example: Piano Extraction from Full Mix

SAM Audio extraction using the prompt 'piano' on a complex classical passage.

Toggle versions while playing to compare.

Net Win Rate of SAM Audio vs. SoTA Separation (source)

How to Use the Universal Stem Splitter

1

Choose Separation Type

Describe any custom instrument or use a preset to isolate with AI.

2

Import Your Audio

Simply drag and drop your audio file or click to browse. We support all major formats including MP3, WAV, FLAC, and more.

3

Download & Create

Our GPU infrastructure processes your audio in seconds. Download individual stems and unlock endless possibilities.

For Every Audio Professional

Music Producers & Engineers

Gain unprecedented control during mixing and mastering. Isolate problematic frequencies, rebalance existing mixes, or extract inspiration from reference tracks with surgical precision.

DJs & Remix Artists

Create clean acapellas for mashups, extract instrumentals for seamless transitions, and break down tracks into building blocks for your next killer set.

Karaoke & Content Creators

Generate high-quality instrumental tracks for karaoke, isolate dialogue from background music for videos, or remove unwanted elements from podcast recordings.

Musicians & Students

Learn songs by isolating individual instruments, practice along with drumless tracks, or transcribe complex parts by soloing specific elements.

Want to go further? Restore your stem from mp3 to 20Khz WAV

Stem extraction often contains mp3 artifacts and is limited to 16Khz. Improve the resulting stem using our audio restoration model.

Spectrogram

Example: stem of a mp3 music generated with Demucs

Frequently Asked Questions

Ready to Isolate Your Tracks?

Experience the best AI stem separation technology.
Clean, fast, and easy to start.

Start Extracting Stems