Eleven v4
ElevenLabs' new flagship voice model: expressive, emotional speech in 90+ languages, with a fast Turbo variant.
Quality
Modality
audio
Runnable
Cloud Only
Access
closed
Fabian's Take
"ElevenLabs is still the best text-to-speech I've used. Since v3 I can just type [laughs] or [whispers] into the text and it performs it. It feels like directing an actor."
Eleven v4 is ElevenLabs’ flagship text-to-speech model since 28 September 2026. It comes in two versions: v4 for the most expressive delivery, and v4 Turbo, which answers in about 100 milliseconds and is now the default for ElevenLabs’ voice agents. Eleven v3 is still available as the previous generation.
Emotional control
Many models produce clear, understandable speech. ElevenLabs is good at pacing and emotion: it reads the context of the text and adjusts the delivery. Inline audio tags, introduced with v3, let you direct the performance line by line instead of regenerating until you get lucky.
Through the API, v4 costs $0.08 per 1,000 characters and v4 Turbo half that. In the app you pay with a monthly plan, from $6 for Starter to $99 for Pro.
The Verdict
Best for: Voiceovers, audiobooks, emotional dialogue, and voice agents (v4 Turbo).
Pros
- Lifelike emotional delivery
- Inline audio tags like [laughs] or [whispers] for precise direction
- Over 90 languages
- A Turbo variant with about 100 ms latency for live conversations
Cons
- Can get expensive for very long-form content
Specs
- Pricing API: $0.08 per 1,000 characters (v4 Turbo $0.04) · app plans from $6/month
- Cost Tier moderate
- ⚡Speed Tier fast
- License Proprietary
