Generating music using AI in Copilot just got even better


Suno AI in Copilot

Sabrina Ortiz/ZDNET

Copilot is a one-stop shop for most of your AI needs: text, image, and even music generation are all available in the service. If you’re a Copilot user, you should know that Copilot’s Suno AI text-to-music-generating plug-in, which first appeared in December 2023, got an upgrade last week.

The tidbit comes from Microsoft CVP for Search and AI Jordi Ribas, who shared via X (previously Twitter) that Microsoft recently rolled out Suno’s new v3 model in the Copilot plug-in. The new model lets users produce better-quality music, with the option of generating more styles, and genres, according to Ribas.

When Suno unveiled its v3 model in March, the startup claimed that v3 was capable of producing “radio-quality music.” Other improvements included “improved prompt adherence,” which enables the more graceful endings Ribas mentioned, as well as fewer hallucinations. 

Also: Photoshop for music? Adobe unveils new AI tools for audio

To try the model yourself, all you have to do is visit Copilot, click on the “Plugins” option next to the chat, and toggle Suno on. You can then ask Copilot to create a song by describing what you want with as little or as much detail as you’d like. As seen in the demo above, Copilot will use Suno to generate the lyrics and then the song, which you can listen to, download, and share. 

Audio-generating AI models have been getting a lot of attention lately. Just last week, OpenAI shared updates to its Voice Engine, a voice cloning AI model, and Stability AI unveiled its most advanced audio model, Stable Audio, which essentially does the same thing that Suno AI does.

Source link