ElevenLabs previously provides AI-produced versions of human voices and music. Now, it will let folks produce sound consequences for podcasts, movies, or online games, too. The new Seem Consequences resource can create up to 22 seconds of appears based on person prompts that can be blended with the company’s voice and songs system, and it presents users at the very least four downloadable audio clip options.
The organization states it worked with the stock media platform Shutterstock to develop a library and educate its model on its audio clips. Shutterstock has accredited its material libraries to quite a few AI providers, together with OpenAI, Meta, and Google.
Sound Outcomes is free to use, but paid out tiers can use the generated audio clips with professional licenses, though free users “must attribute ElevenLabs by which includes ‘elevenlabs.io’ in the title.” ElevenLabs end users have a established character rely limit when writing prompts, with no cost users acquiring 10,000 figures per thirty day period. For Audio Results, ElevenLabs says on its FAQs web site that it will get 40 characters per next from the allotment if people set the audio clip duration themselves. If using the default audio duration, each individual prompt request will be billed 200 people.
Libraries with seem result clips now exist in the sector for creators, filmmakers, and video game builders. But in some cases, these can be costly or have trouble surfacing just the right sort of sound. ElevenLabs says in its weblog publish that it developed Sound Outcomes “to create prosperous and immersive soundscapes quickly, affordably and at scale.”
Other AI builders are also establishing their have textual content-to-seem generators. Steadiness AI introduced Stable Audio last 12 months, which creates audio clips of music and audio effects, and Meta’s AudioCraft versions make normal seem (believe qualifications noises like wind or visitors).