ElevenLabs launches Eleven v4 and v4 Turbo, topping Artificial Analysis TTS Arena
testingcatalog · x · 2026-09-29
ElevenLabs has released Eleven v4 and Eleven v4 Turbo, with v4 taking the #1 spot on the Artificial Analysis voice TTS Arena.
- Eleven v4: quality-focused, built on a new architecture that reads scripts with awareness of speaker, context, and intended tone. Supports 90+ languages, multiple speakers, sound effects, and inline natural-language audio tags like [laughs], [whispers], [door slams] (SSML break tags are disabled), with more reliable tag adherence than v3.
- Context stitching keeps pacing consistent across long-form projects; each generation supports up to 10,000 characters, with improved speaker stability.
- v4 Turbo: built for real-time voice agents and the API, with 100 ms median inference latency and 150 ms median time to first speech, available via ElevenAgents.
Both models are available through ElevenLabs products and the ElevenAPI.
Related event: ElevenLabs Launches Eleven v4 and v4 Turbo, Topping TTS Arenas(18 posts)→
More from Multimodal
- Hyperpop MV made with Claude Code: ~4k lines of code redrawing every frame — repligate · 2026-09-29
- Eleven v4 tested with the author's own voice, nearly removing her accent — ciguleva · 2026-09-29
- ElevenLabs launches Eleven v4, its fastest and most emotive AI voice model yet — Polymarket · 2026-09-29
- Physics-simulated instruments played by an LLM: full rehearsal finally succeeds — alexanderchen · 2026-09-29
- AI video tip: import a movie scene's camera move, replace everything else — LearnWithBishal · 2026-09-29
- Villain-boss POV video made with Magnific and Seedance 2.5, prompt shared — charis_ai · 2026-09-29