ElevenLabs launched a text-to-sound effects AI audio model

ElevenLabs, the company behind the pioneering multilanguage text-to-speech and voice generation platform, has launched a text-to-sound effects model it first announced in February. The new technology lets users generate varied audio content, from sound effects and short instrumental tracks to soundscapes and varied character voices—simply by providing a text prompt describing the idea.

The Sound Effects model was built in partnership with Shutterstock, which enabled ElevenLabs to tap into Shutterstock's extensive licensed sound library during the fine-tuning stage of the model's training. Commenting on the collaboration, Shutterstock CEO Aimee Egan remarked on how Shutterstock's ethically sourced data and ElevenLabs' cutting-edge technology had come together to build "a true market first" enabling the community to create the most diverse projects.