Eleven Multilingual v2
ElevenLabs' flagship text-to-speech model across 29 languages
Eleven Multilingual v2 is the speech synthesis model ElevenLabs shipped when it moved its text-to-speech product out of beta, in August 2023. It generates natural-sounding speech across 29 languages, including Chinese, Japanese, Korean, Arabic, Hindi, and most major European languages, while preserving a cloned voice’s character between languages. Compared to ElevenLabs’ faster Flash and Turbo models, Multilingual v2 trades speed for quality: it’s meant for narration, dubbing, audiobooks, and other content where the audio gets published rather than streamed live in a conversation. It remains one of ElevenLabs’ most widely used models for voice cloning and dubbing workflows even after the company released Eleven v3.