ELYUMCREATE. EDIT. MOVE.ModelsPricingCalculatorSign inStart creating

Eleven v4AI audio generator

Expressive speech with audio tags and the full voice library. Generate, listen and download. Failed generations are refunded.

Text to speech
Your audio, ready to use. See the quote before generating. Successful audio is charged on render.
Open Audio studioSee pricingSee your exact quote before generating
Developed by ElevenLabs
Max resolutionPER 1,000 CHARACTERS

Turn your script into spoken audio with Eleven v4. Choose an available voice or search ElevenLabs’ public voice library. Listen to a sample before generating your own take.

Use inline audio tags for delivery, then choose Creative, Natural or Robust stability. Each request accepts up to 10,000 characters. The base rate is 5.6 credits ($0.056) per 1,000 characters, 30% below the comparable Fal speech model. Charges scale with script length and round to whole credits, with a one-credit minimum. Failed generations are refunded. This model requires eligible paid access on Elyum.

Eleven v4 pricing on Elyum

Billing unitCredits / 1,000 charactersUSD
per 1,000 characters5.6$0.056

Base rates are 30% below the comparable Fal speech model. Charges scale with script length and round to whole credits, with a 1-credit minimum.

1 credit = $0.01. Audio is charged on successful render; failed generations are refunded. Full plan details on the pricing page.

Why Eleven v4 is good

Find a voice by sound

Search available voices and the public library. Filter the library by language and gender, read voice descriptions, and play the supplied previews.

Direct the delivery

Use inline audio tags for delivery, then choose Creative, Natural or Robust stability.

Control the script

Write up to 10,000 characters. Set speed and text normalization. Add surrounding passages for context.

Download your take

Play generated speech in the Audio tab, seek through it and download the MP3. Choose from three MP3 quality settings before generating.

Why generate with Elyum?

  • Failed audio generations are refunded.
  • Every top audio model in one place, one balance.
  • Base rates 30% below comparable Fal speech models, before credit rounding.
  • Download completed audio directly from the studio.
  • Works from the studio, or from Claude, ChatGPT and Cursor via MCP.
  • Free plan — our algorithm may grant you 100 credits, no card required.

How it works

  1. Open Audio, select Speech and choose Eleven v4.
  2. Browse or search voices. Play a preview, then select the voice you want.
  3. Enter your script within the 10,000-character limit. Adjust delivery and voice settings.
  4. Generate your take, play it and download the MP3. Successful renders use credits; failures are refunded.

What people make with it

  • Narration for a product walkthrough or tutorial.
  • Spoken lines for a video scene or game character.
  • Podcast introductions and short scripted segments.
  • Audio versions of written announcements or lessons.

Frequently asked questions

Is Eleven v4 free to try?
Browsing voices and listening to their supplied samples is free. Generating your own speech requires eligible paid access and credits. Audio is charged on successful render; it does not use the video and image Keep/Kill workflow.
How much does Eleven v4 cost?
The base rate is 5.6 credits ($0.056) per 1,000 characters. The final charge scales with script length and rounds to whole credits, with a one-credit minimum. Failed generations are refunded. Check the Generate button for your exact quote before submitting.
Which voice controls are available?
Use inline audio tags for delivery, then choose Creative, Natural or Robust stability. You can also set speed, MP3 quality and text normalization. Surrounding passages can provide context. The Audio tab shows the controls supported by the selected model.
Do I need my own ElevenLabs account?
No. Choose voices and generate speech inside Elyum using your eligible paid access and Elyum credits.
Can I download and reuse a take?
Yes. Completed speech is available as an MP3 from the Audio tab and your Library. Use Copy on a take to restore its script, voice, model and settings for another generation.
Does this include voice cloning or dubbing?
This workflow generates speech from text using existing voices. Voice cloning, voice design, dubbing, transcription and voice conversion are separate tools and are not included here.

More models to try

Eleven v4 TurboExpressive speech with audio tags and the full voice libraryfrom 1 creditsEleven v3Expressive speech with audio tags and the full voice libraryfrom 1 creditsEleven v3 ConversationalExpressive speech with audio tags and the full voice libraryfrom 1 creditsEleven Multilingual v2Natural speech with the full voice libraryfrom 1 creditsEleven Flash v2.5Natural speech with the full voice libraryfrom 1 creditsEleven Turbo v2.5Natural speech with the full voice libraryfrom 1 credits

Create everything. Keep the best.

Eleven v4 and 50+ other top models on one balance. Generate and download audio from one studio.

Open Audio studio