Seedance 2.5 is here. Every model is included on every plan.
PricingLog inSign up
VoiceElevenLabs

ElevenLabs V3

ElevenLabs V3 is the expressive ElevenLabs model, capable of multi voice dialogue in a single generation. Reach for it when a script has more than one speaker or needs real emotional range.

Voices302 available
Make with ElevenLabs V3 →View pricing

Added July 2026. Included on every plan.

Why this one

What makes ElevenLabs V3 different

  1. Multiple voices in one renderA two or three hander comes back as one coherent take rather than as separate clips to assemble.
  2. Wider emotional rangeIt leans further into emotion than the classic model, which suits drama and character work.
  3. The same large voice libraryHundreds of voices to cast from, so a scene is cast rather than settled for.
  4. Per character pricingCost follows script length, so a scene is easy to budget.
  5. Good scene timingTurn taking between speakers lands with realistic gaps.
  6. MultilingualThe expressive model carries across languages too.

Use cases

Best for

  • Character dialogue
  • Ad scripts
  • Audio drama
  • Game lines
  • Explainers with two hosts
  • Emotional narration

Technique

Prompt tips

  1. Write it as a scriptSpeaker labels and lines. The model is built for dialogue, so give it dialogue shaped input.
  2. Direct emotion in the line, not around itWord choice and punctuation carry more than a stage direction will.
  3. Cast deliberatelyTwo voices that contrast in pitch and pace read as two people; two similar voices read as one confused one.
  4. Keep scenes shortRender a scene at a time so a single bad line does not cost you the whole take.
  5. Use ellipses and dashes sparinglyHeavy punctuation produces heavy pauses. A comma is usually enough.

Mechanism

How ElevenLabs V3 works

You supply a script, optionally with more than one speaker, and the model renders the whole exchange in one pass so that timing and reactions between voices are generated together rather than edited together. Billing is per thousand characters of script, so a longer scene costs proportionally more.

Getting started

How to use it on Dolly

  1. Pick ElevenLabs V3Choose it in the voice studio, or just ask the agent for it by name.
  2. Write the promptDescribe what you want, in as much detail as the brief deserves.
  3. GenerateThe exact credit cost sits on the button before you press it, and a failed run refunds automatically.

Pricing

What it costs

Every generation shows its exact credit cost on the button before you run it.

per 1,000 characters
20 credits

One run is about 1/63 of a Starter month.

Failed generations refund automatically. See plans

Capabilities

Specs

Voices
302

Questions

ElevenLabs V3 FAQ

Can it really do two voices at once?
Yes. Multi voice dialogue in a single generation is the reason to pick this tier over the classic one.
How many voices are available?
Hundreds, all listed on this page, and you can audition any of them in the voice studio before spending a credit.
How is it priced?
Per thousand characters. The table on this page comes from the registry.
V3 or Multilingual V2?
V3 for dialogue and emotional range, since it renders more than one speaker in a single take. V2 for long, even narration where consistency matters more than performance.
Can I control who speaks when?
Write the script with clear speaker turns. The structure of your text is the direction.

More like this

Other voice models

ElevenLabs Multilingual V2ElevenLabs Multilingual V2 is the classic ElevenLabs voice model, with the largest voice library on Dolly and consistently lifelike delivery.Gemini 2.5 Pro TTSGemini 2.5 Pro TTS is Google's studio quality speech model, with a curated voice set and multi speaker scene support.KokoroKokoro is a fast, open source speech model with twenty American voices and the lowest price per word on Dolly.

Ready to make with ElevenLabs V3?

Every model is included on every plan. The price sits on the button before you run it.

Start making →