rvachev.orgEN / RU / 🤖
← Back to essays
· Essay · 1 min

Microsoft Introduces: Stable Diffusion for Sound

Microsoft's new tool allows you to edit and generate sounds using text prompts.

Microsoft's new tool allows you to edit and generate sounds using text prompts. Examples include adding sound to a track, removing voice, replacing laughter with trumpet sound, inpainting (coloring sound), and Super-resolution. All this is achieved through text prompts. Are sound operators following in the footsteps of copywriters?

Examples of prompts:

- Add a car horn honks several times loudly
- Replace laughter to trumpet
- Perform Super-resolution

Demo link: https://audit-demo.github.io/

Paper: https://arxiv.org/abs/2304.00830

Microsoft Introduces: Stable Diffusion for Sound — illustration

Newsletter

Get new posts by email

A weekly digest of what I write about AI, software and products. Unsubscribe anytime.