ElevenLabs ships Eleven v4 and low-latency Turbo for expressive speech

ElevenLabs 发布最富表现力的 Eleven v4 及低延迟 Turbo 版

ElevenLabs released the text-to-speech model Eleven v4 and a low-latency version, Eleven v4 Turbo, with stronger emotional expression, speaker consistency and multi-speaker dialogue. Both support more than 90 languages, and Instant Voice Clones need only 10 seconds of audio. Turbo's median inference latency is about 100ms, with about 150ms from request to audible speech.

Why it matters: ElevenLabs pairs emotional control with low latency in Eleven v4 Turbo, so voice agents can answer fast and still sound natural.