A scene's sounds from one prompt
ElevenLabs, the U.S. voice AI company, on Oct. 29 (Oct. 30 Japan time) released "SFX v3," a new version of its AI model that creates sound effects from text prompts.
When a single prompt describes multiple sounds, more of them are now actually produced. Unwanted noise that occurred in sounds such as bells, chimes and whistles has also been reduced. The company said it also improved results for prompts that include music or vocals.
Claims to outperform the previous version
ElevenLabs said the model outperformed the previous version, SFX v2, in internal evaluations, but did not provide scores. The announcement also does not include comparisons with other companies' models, the length of sounds it can create or pricing.
It is available in the sound effects screen and music editing screen of the ElevenCreative production app, in the mobile app and through the developer API. The day before, the company also released "Cue," a model that aligns audio and text character by character.
