Eleven Realtime Translation
Translate live speech into spoken audio in more than 90 languages through an enterprise WebSocket API.
Generate sound effects, layered audio scenes and musical or vocal sounds from text descriptions.
Visit website
ElevenLabs SFX v3 converts text descriptions into sound effects, including prompts containing multiple events, music or vocals. It supports sound design workflows that describe individual sounds and their surrounding scene.
Eligible users can access it through ElevenCreative’s Sound Effects, Flows and Studio, or through the Sound Effects API. Access is limited and may vary by region, so account eligibility is a key selection consideration.
Describe several sounds or actions within one prompt, including combinations with musical or vocal elements.
Generate repeating effects and ambience with a seamless loop.
The API accepts a specified duration from 0.5 to 30 seconds, or automatic duration based on the prompt.
Adjust prompt influence between 0 and 1. Higher settings increase adherence to the description while reducing variation between generations.
Use the existing Sound Effects endpoint with model_id set to eleven_text_to_sound_v3. The default remains eleven_text_to_sound_v2.
Explore Foley and environmental sounds that support the action in an edit.
Prototype repeating background textures for locations and interactive scenes.
Compare descriptions combining everyday sounds with musical or vocal details.
Offer text-driven sound creation within an application after confirming model access.
ElevenCreative offers a free tier and paid subscriptions with monthly credits, with Sound Effects among its platform features. SFX v3 access remains conditional. Consult the ElevenCreative and API pricing pages for current charges.
Check current pricing ↗No. SFX v2 remains the default. Select eleven_text_to_sound_v3 explicitly and confirm that the account has access.
Yes. SFX v3 handles descriptions containing musical sounds or vocals, including prompts combining several sound elements.
Review the generated audio, adjust the description or settings and regenerate. The web workflow supports reviewing and downloading results.