Skip to content
ElevenLabs Multilingual v2 logo

MODELS

ElevenLabs Multilingual v2

ElevenLabs' flagship TTS — most natural English voices, 29 languages.

elevenlabselevenlabs-tts

Specs

Modalities
text, audio
Tool use
Vision
Streaming
License
proprietary
Released
2023-09-14

Pricing

ElevenLabs Multilingual v2 (September 2023) is the company's flagship text-to-speech model covering 29 languages including Mandarin, Cantonese, English, Japanese, Korean. Voice cloning from 1-minute samples, emotion control, streaming output. Subscription tiers from $5/mo (Starter, 30K chars) to $330/mo (Business, 11M chars + commercial); enterprise tiers add SLA. Used widely in audiobooks, podcasts, video voiceover, and conversational agents.

Editor's verdict

The category-defining English TTS — beats OpenAI's tts-1, Azure Neural TTS, and Google Cloud TTS on naturalness in most blind tests. Mandarin and Cantonese support is real but less polished than the English ones — for Chinese-first products, MiniMax abab or iFlytek Spark voice often sound better. Voice cloning is genuinely 'spooky good' from a 1-minute sample, which is also a content-policy concern; ElevenLabs has progressively tightened verification.

Reviews

No reviews yet. Be the first.

Last updated: 2026-04-29

We use cookies

Anonymous analytics help us improve the site. You can opt out anytime. Learn more