Elevenlabs releases an open-source demo to add sound effects for videos and announces new partnerships

Elevenlabs recently built and open-sourced a demo to showcase its Texts to Sounds Effects API. The demo's code is accessible on GitHub, and the demo can be accessed as a web app where users can upload any video to add AI-generated sound effects. The tool extracts four frames at 1-second intervals (client-side), sends the frames and a prompt to GPT-4o to create a custom prompt for the Text-to-sound effects API, uses the API to create a sound effect, and combines it with the input video using ffmpeg.wasm to create a single downloadable file.

In addition to the demo release, Elevenlabs detailed some recent partnerships and collaborations. These include partnering with PocketFM to launch AI Audio Series, a tool that allows writers to transform their narratives into appealing audio series; supporting Infer.so's multimodal voice bot by lending it authentic, user-friendly sounding voices; collaborating with the creative production firm Tool to finish a mixed media campaign for Under Armour; and playing a key role in AnyTopic's audiobook creation pipeline.

As generative AI for media generation continues to gain traction, Elevenlabs has demonstrated that it is determined to remain an industry leader by continuing to develop its technology and support creators across several industries.