Bottom line: Eleven v4, released September 28, 2026, is ElevenLabs’ most emotive text-to-speech model yet — 90+ languages, inline audio tags instead of SSML, instant clones from 10 seconds of audio, and Professional Voice Clones restored. On the API it is not free-tier cheap: v4 lists at $0.08 per 1,000 characters and v4 Turbo at $0.04, so Turbo is the rational default for real-time agents.
ElevenLabs launches Eleven v4 — its most emotive voice model yet, plus a Turbo for real-time agents
Published September 28, 2026. ElevenLabs has released Eleven v4, a text-to-speech model built on an entirely new architecture, alongside Eleven v4 Turbo, a low-latency variant aimed at voice agents and real-time use. Both are available now in ElevenAgents, ElevenCreative, and via API. For the full platform picture, see our ElevenLabs review.
What’s new
- A new architecture designed to interpret tone, pacing, emotion, character, and context — generating speech that can sound dramatic, tender, urgent, comedic, or conversational while holding the speaker’s identity.
- Inline audio tags replace SSML. You direct delivery in plain language — tags like [laughs], [said angrily in French accent], even [light rain] or [phone buzzing] — and the model follows them more accurately than its predecessors. Classic SSML break tags are disabled in v4 — a documented gotcha for anyone migrating v3 scripts.
- 90+ languages, a 10,000-character limit per generation (double v3), and instant voice clones from as little as 10 seconds of audio. Professional Voice Clones, missing from v3, are supported again in v4.
- Better speaker-identity capture keeps voices consistent across agent conversations, audiobooks, and ads; scene-level context makes multi-line dialogue sound natural rather than stitched together.
- Improved IPA phoneme support for reliable custom pronunciations. Turbo’s median time to first speech is about 150 ms in ElevenLabs’ own testing, which is why it, not the full model, is pitched at live agents — a trade-off we break down in ElevenLabs vs Cartesia.
Pricing check
ElevenLabs’ API pricing lists Eleven v4 at $0.08 per 1,000 characters and Eleven v4 Turbo at $0.04 per 1,000 characters — Turbo at half the price, which matters at agent scale. Subscription tiers still gate total credits, so compare the models in context in our ElevenLabs vs Murf vs Speechify comparison and our ElevenLabs vs PlayHT comparison, or browse the best ElevenLabs alternatives by use case.
Why it matters
Voice is where “good enough” AI audio becomes genuinely usable — for faceless YouTube channels, audiobooks, and voice agents, the difference between flat narration and directed performance is the whole product. On the published specs and pricing above, v4 looks like the release that raises the bar for everyone else. This coverage is research-based on ElevenLabs’ launch and pricing material; we make no hands-on testing claim here.
Sources: ElevenLabs, “Introducing Eleven v4” and ElevenLabs API pricing (September–October 2026); Unite.AI; TechCrunch launch coverage.