Understand the technology behind AI-powered stem separation and learn how to get the most out of Vocal Division.
Vocal Division uses Demucs v4, a state-of-the-art deep learning model trained on thousands of songs. The AI analyzes frequency patterns, stereo information, and temporal structure to isolate different instruments with studio-quality precision.
Isolate lead and backing vocals with crystal-clear separation
Extract percussion elements including kicks, snares, and cymbals
Separate low-frequency instruments and basslines
Everything except vocals — perfect for karaoke tracks
Drag and drop any audio file (MP3, WAV, FLAC) or paste a YouTube link. Maximum 10 minutes, up to 100MB. The AI works with any genre — from hip-hop to classical.
Choose between 2-stem (vocals + instrumental) or 4-stem (vocals + drums + bass + instrumental). Processing takes 30-90 seconds depending on song length.
Preview each stem with the built-in player. Use the mixer to adjust volume levels and mute tracks, then export your custom mix as high-quality WAV or MP3.
Volume control, mute buttons, and real-time preview for each stem
Automatic tempo analysis using Essentia AI (±2 BPM accuracy)
Musical key identification for harmonic mixing and remixing
Download individual stems or your custom mix in WAV or MP3
Create acapellas and instrumentals for live mixing and mashups
Sample vocals and drums, study arrangements, remix tracks
Practice with isolated vocals or create karaoke backing tracks
Extract music elements for YouTube videos and podcasts