- Home
- Stem Separation Software
- UnMixLabs
UnMixLabs
What it is
UnMixLabs is a web service and API that separates finished audio into vocal and instrument stems. It accepts MP3, WAV, FLAC, M4A, OGG, MP4, and MKV files, with 4-stem, 6-stem, 53-stem, and vocal-isolation modes. Stem categories include vocals, drums, bass, guitar, piano, instrumental, and other parts. Every job includes BPM, musical-key, and chord-progression analysis. Optional tools provide vocal dereverb, stem-to-MIDI conversion, beat-grid export, synced lyrics, karaoke video, and audio cleanup presets. Outputs can include MIDI, beat-grid TSV, REAPER sessions, lyric SRT files, karaoke MP4s, and WAV, MP3, or FLAC audio. Batch processing is available, and Studio includes API access for automation. The Free plan allows 10 processing minutes per month, one 100 MB file at a time, 4-stem separation, and 24-hour file retention. Pro and Studio add larger files, batch sizes, processing allowances, longer retention, and priority queues. The pricing page says the service is free-only during beta while payment integration is prepared.
For and against
For
- Accepts seven listed input formats
- Offers 4-, 6-, and 53-stem modes
- Includes musical analysis with every job
- Optional MIDI and karaoke outputs
- Free plan available
Against
- Free plan limits processing to 10 minutes monthly
- Free files retain for 24 hours
- Paid subscriptions are not currently available during beta
The verdict
UnMixLabs combines stem separation with analysis and optional production outputs. Its free beta access covers basic jobs, while planned paid tiers describe larger files, batches, retention, and API use.
Who it suits
It suits musicians, producers, podcasters, and studios needing separated stems or batch audio processing. API access is aimed at automation workflows.
Plans and prices
Free
10 minutes processing/month · 4-stem split · 100 MB max file · 1 file at a time
Read 7 Oct 2026 · unmixlabs.com
Pro
300 minutes/month + rollover · all 4 models · 1 GB max file · up to 20 files per batch
Read 7 Oct 2026 · unmixlabs.com
Studio
1500 minutes/month + rollover · all 4 models · 2 GB max file · up to 50 files per batch
Read 7 Oct 2026 · unmixlabs.com
On the record
- Analysis
- The service says every job includes free BPM, musical-key, and chord-progression analysis.unmixlabs.com · read 7 Oct 2026
- API
- The site describes a partner API and lists API access for batch automation in the Studio plan.unmixlabs.com · read 7 Oct 2026
- Audio inputs
- The homepage lists MP3, WAV, FLAC, M4A, OGG, MP4, and MKV as accepted formats.unmixlabs.com · read 7 Oct 2026
- Audio tools
- Optional tools include vocal dereverb, stem-to-MIDI, beat-grid export, synced lyrics, karaoke video, and audio cleanup presets.unmixlabs.com · read 7 Oct 2026
- File retention
- The privacy page says free-tier uploads are deleted when processing finishes and stems are deleted 24 hours after the job completes.unmixlabs.com · read 7 Oct 2026
- Integrations
- The site says optional outputs include MIDI, beat-grid TSV, REAPER sessions, lyric SRT files, and karaoke MP4s.unmixlabs.com · read 7 Oct 2026
- Intended users
- The Studio plan is described as intended for producers, podcasters, and studios processing volume.unmixlabs.com · read 7 Oct 2026
- Plan availability
- The pricing page says the service is free-only during beta while the maker firms up payment integration, and invites users to join the first cohort for paid subscriptions.unmixlabs.com · read 7 Oct 2026
- Privacy
- The privacy page says uploaded audio is not used to train models by default and that the service does not collect email addresses on the free tier.unmixlabs.com · read 7 Oct 2026
- Purpose
- UnMixLabs separates finished audio tracks into vocals, drums, bass, and other instrument stems.unmixlabs.com · read 7 Oct 2026
- Security
- The privacy page states that public connections use HTTPS and that API-layer job isolation limits access to a user's own jobs.unmixlabs.com · read 7 Oct 2026
- Separation modes
- The site offers 4-stem and 6-stem separation, a 53-stem mode, and vocal-isolation options.unmixlabs.com · read 7 Oct 2026
