OpenAI is reportedly working on a new AI tool that can create music from text or audio prompts, according to The Information. This means users could soon generate full soundtracks or add instruments like a guitar to their own vocals, all powered by AI. The project shows OpenAI’s growing focus on creative tools beyond text and video.
While the exact launch date isn’t known yet, experts believe the tool could either be built into ChatGPT or connected to its upcoming video platform, Sora.
OpenAI Collaborates With Juilliard Students To Train Its Music AI
Reports suggest that OpenAI is teaming up with Juilliard School students, one of the world’s top music institutions, to help annotate and study musical scores.
These annotations are important because they can teach the AI how to understand rhythm, emotion, and harmony, making its music sound more natural and expressive.
This collaboration could give OpenAI’s model a big edge over existing tools by helping it learn from real musicians instead of just raw data.
Sources say these efforts are part of OpenAI’s long-term goal to make AI-generated art feel more “human.”
Expanding OpenAI’s Creative Tools Beyond ChatGPT
OpenAI has been experimenting with sound and music for a while, but most of its recent audio projects focused on speech, like text-to-speech and voice cloning.
Now, the company seems ready to move into the music space, competing with platforms such as Google’s MusicLM and Suno AI, both known for text-to-music generation.
If this new tool succeeds, it could completely change how creators make soundtracks, jingles, and background scores. Musicians, filmmakers, and even students could make professional-quality music without needing instruments or complex editing software.
Meanwhile, OpenAI’s video app Sora is also getting major updates. Bill Peebles, head of Sora, said the platform will soon include AI cameo creation, allowing users to feature pets, toys, or virtual characters in their videos, a big step toward more interactive storytelling.
