OpenAI is developing music generation tool

OpenAI is developing a tool for music generation based on text and audio prompts. This is reported by The Information citing sources. Such a tool could be used to add music to existing videos or to add guitar accompaniment to a vocal track.

It’s still unclear when OpenAI plans to launch the product and whether it will be available as a separate application or integrated with ChatGPT and the Sora video application. One of the sources told The Information that OpenAI is working with students from the prestigious Juilliard School to have them annotate scores. This is a way to obtain training data.

Curious detail: OpenAI already launched generative music models in the past, but that was before ChatGPT. And recently the company focused on audio models for text-to-speech and speech-to-text conversion. Now it’s returning to music.

In this market OpenAI already has competitors – Google and Suno. It turns out that after text, images, video and voice, AI is mastering another creative field. And music becomes the next battlefield of generative models.

A question remains: if Juilliard School students annotate scores for AI training, aren’t they preparing their own replacement? This is the classic automation story – people help create technology that then takes their place.