OpenAI is reportedly working on a new AI-powered tool that can generate music using both textual prompts and audio recordings, according to sources cited by The Information. The company aims to give users the ability to create musical accompaniment for videos, or even add instrumental backing to vocal tracks.
Details about the release date or whether the product will be a standalone app or integrated into ChatGPT and OpenAI’s video model Sora have not yet been disclosed.
According to insiders, OpenAI is collaborating with students from the Juilliard School, one of the world’s leading music academies. These students are helping to annotate musical scores to teach the AI to understand composition, structure, and musical arrangements.
OpenAI already has experience in audio generation, but after focusing on language models post-ChatGPT, this project could mark a major return to music-focused AI development.
The company faces competition from Google and the startup Suno, which have been advancing text-to-music AI tools, with Suno even integrating its technology into Microsoft Copilot. Experts say that combining OpenAI’s text and audio capabilities could produce a tool capable of competing with leading AI music platforms, expanding OpenAI’s ecosystem in creative technology.
