Eleven v4’s Provider Voice Elo of 1,319 surpasses Cartesia’s Sonic 3.6 and Google’s Gemini 3.8 Flash TTS on the Artificial Analysis arena leaderboard. The new text-to-speech model from ElevenLabs also earned a pronunciation robustness score of 91.7%, the highest measurement ever recorded by the analysis firm. It is currently the only model to score above 93% in three of the four pronunciation categories, including the expanding shorthand test.
The update expands language support to more than 90 languages, up from 70 in the previous Eleven v3. Generation speed has increased to 73.4 characters per second from 42.5 characters per second in v3, while the cost is $80 per 1 million characters. While topping the provider rank, Eleven v4 is No. 2 in the controlled voice category behind Alibaba’s Qwen-Audio-3.1-TTS-Plus.
Key sources
- SOURCEmarketbrief.now