ElevenLabs says its new model, Eleven v4, can move a voice from dramatic to conversational while keeping the speaker’s identity.
Creators need to know if that change matters for their work and which product and plan to pick. This article covers what Eleven v4 changes, which product fits each kind of work, and what the plans cost.
What Eleven v4 brings
ElevenLabs says Eleven v4 can shift speech from dramatic to conversational while keeping the speaker’s identity. It also adjusts tone and pacing to the context.
Cloning is stronger too, for text-to-speech. For conversational agents, Eleven v4 Turbo is the low-latency version, with a median inference latency of about 100 milliseconds.
Eleven v4 and Eleven v4 Turbo are both available in ElevenCreative, ElevenAgents, and through the API. Your work decides the product, not the model.
To test either one, it is enough to start with a free ElevenLabs account. A free account can generate with both models before you commit to a paid plan.
Voice cloning, control and languages
An Instant Voice Clone can capture a voice with high fidelity from 10 seconds of audio. For the highest fidelity, use Professional Voice Clones.
Eleven v4 keeps a speaker’s identity more reliably across generations, dialogue, narration and regenerated lines. For a character or narrator in a long project, that helps the voice stay consistent when you redo a line.
Three controls change how you direct a voice:
- Audio tags: v4 follows tags like [laughs] or [light rain] more closely than earlier models.
- Phonemes: support for the International Phonetic Alphabet (IPA) has improved, so custom pronunciations hold up better.
- Request stitching: chained generations for longer content are more reliable in v4.
Which product fits and what it costs
For voiceovers, ads, podcasts, audiobooks and videos, use ElevenCreative. For customer service, sales or operations, use ElevenAgents, which lets you set up, deploy and monitor conversational agents.
Pick Turbo when the voice must reply in real time, such as on a live call with an agent. For recorded content, its faster replies matter less.
Start with how much audio you generate each month, then compare the plans:
- Free: $0 per month, with 10,000 credits included.
- Free: unused credits do not roll over to the next month.
- Paid plans: compare each plan’s monthly price on the pricing page with the credits your projects use.
Eleven v4 matters if your work needs a voice that shifts tone while keeping the same speaker. Choose the product by the job: ElevenCreative for narration and ads, ElevenAgents for support.
Next step: open a free ElevenLabs account and generate the same line with both models. Move to a paid plan only when your monthly audio needs more credits.
Questions and answers
What is Eleven v4 Turbo?
Eleven v4 Turbo is the low-latency version of Eleven v4, built to work with ElevenAgents. Latency is the time the model needs to generate speech: its median inference latency is about 100 milliseconds, or one tenth of a second.
How many languages does Eleven v4 support?
Eleven v4 supports more than 90 languages. A voice recorded in one language can speak any other supported language while keeping its original identity. For a creator, that means one voice can carry a project into several languages.
How much audio does an Instant Voice Clone need?
An Instant Voice Clone needs 10 seconds of audio. That is the amount ElevenLabs gives for a clone with high fidelity, so one short recording is enough to start.
Use a clip with one voice and little background noise, since the clone is built from that recording alone.
Is there a free plan?
Yes, there is a free plan: it costs $0 per month and includes 10,000 credits each month.
Unused credits do not roll over, so they do not carry into the next month. You can test both models on the free plan before you pay.