ElevenLabs introduces Eleven v4 and v4 Turbo voice models
ElevenLabs says v4 lets creators direct speech delivery with inline tags, while v4 Turbo is built for real-time use with a median inference latency of about 100 milliseconds.
TLDR
ElevenLabs introduced Eleven v4 and v4 Turbo, saying the models support 90-plus languages. The company says creators can use inline tags to direct v4’s delivery, emotion and pacing, while v4 Turbo is designed for real-time use. Artificial Analysis ranks Eleven v4 first on its Provider Voice leaderboard and Pronunciation Robustness benchmark, and second on Controlled Voice.
Combined views
1.1M
23 Sources, first seen 5h ago
