Skip to main content
Cartesia Sonic 3 is a low-latency multilingual text-to-speech model with a large catalog of voices across 40+ languages. Voices are referenced by UUID. Pass the voice_id from the tables below as the voice field in your init message.
Many voices support multiple languages. To override the default language for a multilingual voice, set the optional language field to a supported ISO code (for example "language": "es").

Supported languages

English

Spanish

Hindi

French

German

Japanese

Korean

Portuguese

Arabic

Chinese

Dutch

Italian

Polish

Russian

Swedish

Tagalog

Telugu

Turkish

Tamil

Czech

Hebrew

Finnish

Thai

Danish

Bengali

Hungarian

Vietnamese

Bulgarian

Croatian

Georgian

Greek

Gujarati

Indonesian

Kannada

Malay

Malayalam

Marathi

Norwegian

Punjabi

Romanian

Slovak

Ukrainian