ElevenLabs' most emotive text-to-speech voice models yet, with in-script audio direction, voice cloning, and a real-time Turbo variant.
Eleven v4 features
1 / 2Overview
What is Eleven v4 and Eleven v4 Turbo?
Our most expressive models yet
Eleven v4 and Eleven v4 Turbo are ElevenLabs' newest text-to-speech voice models. Eleven v4 is built on an entirely new architecture that reads a script the way a voice actor would — with an exceptionally wide emotional range, multiple speakers, and sound effects across 90+ languages. Eleven v4 Turbo brings the same expressive range to real-time use with about 100–150ms median first-audio latency, streaming text in and audio out, for conversational voice agents.
Direct in-script direction: add tags like [laughs], [whispers], or [door slams] straight into the script and v4 performs them more reliably than v3. Context stitching keeps pacing and delivery steady across scripts of any length, and regenerating a line no longer drifts the vocal identity. Professional Voice Clones return in v4 (absent in v3) with the model's full emotional range; every clone requires verified consent from the voice owner.
ElevenLabs is freemium: Free $0/month (10,000 credits, ~10 minutes of audio, no card required), then Starter $6, Creator $22, Pro $99, Scale $299, Business $990/month, and custom Enterprise. v4 uses the same credit pricing as other TTS models, and a launch promo offers 2x credits for v4 TTS for two weeks on Creator plans and up.
Platforms and languages
Eleven v4 and Eleven v4 Turbo Availability
Platforms
Languages
Capabilities
Eleven v4 and Eleven v4 Turbo Key Features
New emotive architecture
Eleven v4 reads a script the way a voice actor would, with an exceptionally wide emotional range, multiple speakers, and sound effects in 90+ languages.
In-script audio direction
Add direction like [laughs], [whispers], or [door slams] straight into the script; v4 follows tag sequences more reliably than v3, sound effects included.
Long-form consistency
Context stitching keeps pacing and delivery steady across scripts of any length; regenerating a line no longer drifts the vocal identity.
Voice clones return
Instant and Professional Voice Clones work with v4's full emotional range; Professional clones return after being absent in v3, with verified owner consent.
Eleven v4 Turbo for real-time
The low-latency variant for real-time and voice agents: about 100–150ms median first-audio latency with streaming text and audio.
Best for
Who uses Eleven v4 and Eleven v4 Turbo?
Creators and content teams producing audiobooks, narration, dubbing, and ads
Generate expressive, controllable voice audio at scale
Developers building real-time conversational voice agents
Use the low-latency v4 Turbo model via API with streaming
Plans and access
Eleven v4 and Eleven v4 Turbo Pricing
Freemium
Free $0/mo (10k credits). Starter $6, Creator $22, Pro $99, Scale $299, Business $990/mo. 2x v4 credits promo.
Common questions
Eleven v4 and Eleven v4 Turbo FAQs
Does v4 support voice cloning?
Yes — Instant and Professional Voice Clones both work with v4; Professional clones return after being absent in v3, and every clone requires verified consent from the voice owner.
How many languages are supported?
90+ languages, per the official FAQ — built for global audiences.
What is the difference between v4 and v4 Turbo?
v4 is tuned for top-quality finished content; v4 Turbo is the low-latency variant (about 100ms median inference) for real-time voice agents. Both share the same expressiveness and support Professional Voice Clones.
Is there an API?
Yes — streaming and non-streaming endpoints launched with the models, plus TypeScript and Python SDKs.
How is it priced? Is there a free plan?
Same credit pricing as other TTS models on all plans, and there is a free tier with 10,000 credits per month (about 10 minutes of audio).
Reviews
See what the community thinks and share your experience.
—
Based on 0 ratings
Rating distribution
Community reviews
No reviews
No written reviews yet.
