Voice Isolator
Voice Isolator is ElevenLabs’ trained deep-learning model for separating vocal signals from an audio mix. It processes recorded audio rather than generating a new spoken script.
Recording cleanup
The API accepts audio and video files and returns isolated speech. Its intended inputs include interviews and recordings with ambient noise, background music or other interference.ElevenLabs Voice Isolator API capabilities
Practical guidance. Compare the isolated recording with the original before using it for transcription or editing. Removing background sound can also change the sound of the speech you need to preserve.ElevenLabs Voice Isolator API capabilities
Specifications & access
| Detail | Value and conditions |
|---|---|
| Access | API, Provider product |
| Input limits | 500 MB and one hourElevenLabs Voice Isolator API capabilities |
| Music limitation | Not specifically optimized for isolating sung vocalsElevenLabs Voice Isolator API capabilities |
Related models
- Scribe v2
ElevenLabs · Scribe
Batch speech-recognition model for long recordings, structured transcripts and multilingual audio.
Launch date API · Provider product - Multilingual Voice Changer v2
ElevenLabs · Eleven Voice Changer
Voice-conversion model that transfers a recorded performance to another voice across 29 languages.
Launch date not confirmedAPI · Provider product