Eleven Multilingual v2

Audio modelPrevious generationCard summarized

Lifelike, consistent quality speech synthesis model

Source: ElevenLabs Comes Out of Beta and Releases Eleven Multilingual v2 (launch post)

Use it via API

$0.08 / 1K chars

Code example Compare

Where to call it

ElevenLabs APIeleven_multilingual_v2

Supports

Streaming

Full feature support

Good for

Character voiceovers for games and animation
Corporate video and e-learning narration
Multilingual projects with a consistent voice across languages

Replaced by

Eleven Flash v2.5

Compare side by side

Overview

ElevenLabs · 07TTS

Multilingual v2

Lifelike, consistent quality speech synthesis model

Typical price

$0.08 / 1K chars

v2 Multilingual text to speech

Max length: ~10 min (10,000 characters)
Languages: 29 languages
Modes: text-to-speech, streaming
Released: 22 Aug 2023

Eleven Multilingual v2 is a foundational speech model that identifies nearly 30 written languages from the input text and generates speech in them, keeping a synthetic or cloned voice's characteristics, including its original accent, across languages. The current docs call it a previous-generation model, most stable on long-form generation, supporting 29 languages. launch post; Models page

eleven_multilingual_v2

Typical price: $0.08 / 1K chars
Max length: ~10 min (10,000 characters)
Languages: 29 languages
Weights: Closed
Released: 22 Aug 2023

text-to-speech
streaming
WebSocket input streaming
voice cloning (instant and professional)