The American technology company OpenAI is working on a new artificial intelligence model that can create music. The Information reported this, citing sources.
As part of the new project, OpenAI is collaborating with The Juilliard School, one of the most prestigious art and music education institutions in the United States. Together, the company and the educational institution plan to digitally tag musical notes and use them as AI training data.
The new model will allow it to create music “from scratch” or add instrumental accompaniment (for example, guitar) to an existing vocal sound. This technology could allow OpenAI to compete with music generation platforms such as Suno and Udio.
Detailed information about the project has not yet been disclosed. It is unknown whether the new system will be integrated with ChatGPT or Sora or will be presented as a separate application. OpenAI has not made an official comment on the matter.
It should be noted that the company has previously presented music-generating models - in 2019, MuseNet, and in 2020, Jukebox, systems capable of creating instrumental and vocal music using neural networks, were demonstrated.
