AI text-to-speech model

ElevenLabs Eleven v3

Prompt: Voice actor at a studio microphone with pop filter

ElevenLabs' most expressive TTS model for narration, dialogue, and character voices with picker-driven voice selection and voice tuning.

Generate speechExplore outputs
Inputs
  • Text prompt
  • Voice selectionPick from the synced voice library
  • Language hint
  • Voice cloningCustom voices coming soon
Settings
  • Voice
  • Output formatmp3, wav, opus, pcm
  • Voice speed
  • Voice stability
Outputs
  • Audio outputDefault mp3, alternative pcm/wav/opus
  • Voice metadata
  • Character counting
Strengths

Expressive speech that carries intent

Use Eleven v3 when the line needs emotion, pacing, or a specific character feel rather than a flat read.

Rich voice catalog

Pick from synced ElevenLabs voices filtered by language, gender, age, and accent.

Voice settings

Tune stability and speed so the same voice can sound calm, urgent, or natural.

Multiple output formats

Pick mp3 for delivery, wav or pcm when you need to post-process the audio.

Workflow

From script to voiced clip

Treat the text like a performance cue: write how it should be read and then pick the voice that fits.

  1. 1

    Write the line

    Use short sentences, punctuation, and natural phrasing so the voice reads it the way you hear it in your head.

  2. 2

    Pick the voice

    Open the voice picker and filter by language, gender, or age, then play previews before selecting.

  3. 3

    Tune and regenerate

    Adjust stability and speed, or switch voice, until the read matches the scene.

Details

Technical notes

Use these notes as production guidance inside Creativ Studio. Provider availability and exact limits can vary by account, region, and model rollout.

Best for

Expressive narration, ads, and character dialogue

Inputs

Text and a voice selection

Prompt limit

10,000 characters per request

Output formats

mp3, wav, opus, pcm

Prompt recipes

Use these as starting shapes, then swap in your own subject, references, and constraints.

Brand narration

Welcome to Creativ Studio, where creative teams turn prompts into finished assets in minutes, not days.

Character line

Tell him I said hello. And tell him I never forget a face — especially not his.

Questions

How do I pick a voice?

Open the voice picker from the generation settings. Filter by language, gender, age, or accent, and play a preview before selecting.

Can I tune how the voice reads?

Yes. Adjust the voice speed and stability settings to shift the read between calm and expressive.

Which output format should I pick?

Use mp3 for most delivery. Use wav or pcm when you plan to edit the audio in another tool.