Multilingual Voice Changer v2
Multilingual Voice Changer v2 changes the voice in recorded speech while retaining its delivery. It supports 29 languages and can use a cloned or designed output voice.
Preserving a performed line
Voice conversion preserves source timing, pacing and emotion. The original research explains a tradeoff: retaining the source performance too strongly can reduce the target voice’s character, while stronger target identity can weaken the performance.ElevenLabs: original Voice Changer research
Practical guidance. Record the delivery you want before converting it. This is useful when a performer can demonstrate a pause or emphasis more precisely than a text instruction can describe it.ElevenLabs: original Voice Changer research
Specifications & access
| Detail | Value and conditions |
|---|---|
| Access | API, Provider product |
| Model identifiers | eleven_multilingual_sts_v2 |
| API identifier | eleven_multilingual_sts_v2ElevenLabs model catalog and deprecations |
Related models
- Dubbing v2
ElevenLabs · Eleven Dubbing
Audio-to-audio dubbing model that transfers a recorded performance into another language.
Launch date PreviewAPI · Provider product - English Voice Changer v2
ElevenLabs · Eleven Voice Changer
English voice-conversion model for replacing a speaker’s voice in a recording.
Launch date not confirmedAPI · Provider product