OpenVoice

OpenVoice TTS

MyShell.ai's instant voice-cloning model with granular control over style, emotion, accent, and rhythm.

Sign up for 5,000 character limit

Wrap your text in SSML tags for precise control:

<speak><prosody rate="slow">Slow speech</prosody></speak>

Tags the selected model understands — click to drop one into your text where it happens:

This model reads plain text, so inline tags are ignored. For tag-based emotion, switch to an expressive model like Orpheus or Bark.

Define custom pronunciations (word = pronunciation):

-12 +12
0.5x 2.0x
Free with Piper, VITS, MeloTTS
Your generated audio will appear here. Choose a model, enter text, and click Generate.
Audio Generated Successfully
0:00
Download Audio Download .srt Link expires in 24h
Free tier: personal use. Commercial license from $5/mo
Love TTS.ai? Tell your friends!

About OpenVoice

OpenVoice from MyShell.ai is built around a ToneColorConverter on top of a MeloTTS backbone, and its defining trait is granular controllability. It clones a voice from a short clip and then lets you independently tune style, emotion, accent, rhythm, pauses, and intonation, generating speech in multiple languages while preserving the speaker's identity. It also doubles as a voice converter for transforming one voice into another. On TTS.ai it covers English, Chinese, Japanese, Korean, French, and Spanish, accepts up to 5,000 characters, and needs roughly 10 seconds of reference audio to clone.

Best for: Voice cloning with fine-grained style control, voice conversion

Browse all OpenVoice voices

At a glance

Developer
MyShell.ai / MIT
License
MIT
Tier
premium
Speed
medium
Voice cloning
Yes
Languages
English, Chinese, Japanese, Korean, French, Spanish
Max characters
5000

OpenVoice voices

Chinese

Chinese
Premium Neutral

Default

English
Premium Neutral

French

French
Premium Neutral

Japanese

Japanese
Premium Neutral

Korean

Korean
Premium Neutral

Spanish

Spanish
Premium Neutral

OpenVoice TTS — FAQ

OpenVoice lets you control style, emotion, accent, rhythm, pauses, and intonation independently of the cloned voice identity, and it can also act as a voice converter to transform one voice into another.

OpenVoice clones from a short clip, with about 10 seconds of reference audio recommended for good results.

Yes. OpenVoice is MIT-licensed. It supports 6 languages: English, Chinese, Japanese, Korean, French, and Spanish.
← All voices