- OpenAI is developing a generative music tool from text and audio prompts
- The tool may offer multi-vocal track generation and AI-assisted mixing features
- Juilliard students are collaborating to annotate music for AI training data
OpenAI is working on a new generative music tool that can create music from text and audio prompts. This innovative tool is expected to allow users to add original soundtracks or instrumental accompaniments to videos, The Information reported. However, there's no official word yet on when it will launch or whether it will be a standalone product or integrated into existing platforms like ChatGPT or the video generator Sora.
The tool is expected to provide features like multi-vocal track generation and AI-assisted mixing, making it appealing to independent musicians and content creators.
As per the report, the project involves collaboration with students from the prestigious Juilliard School, who are helping to annotate musical scores to provide accurate training data for the AI model. The partnership aims to understand how machines can learn and replicate musical patterns and emotions.
Notably, OpenAI's new tool builds on its previous projects, MuseNet and Jukebox, and promises enhanced user control over musical style, tone, and energy. OpenAI released Jukebox, a music generation tool, in 2020, enabling users to create AI music in various genres like reggae and blues. However, it's no longer actively maintained.
Ethical Concerns
OpenAI's entry into the music generation space is expected to intensify competition with other tech giants like Google and startups like Suno, which already offer generative music solutions. The development of this tool has also sparked discussions about copyright, ethical questions, and creative ownership.
Several voices in the music industry have raised concerns about AI companies not fairly compensating artists, with Paul McCartney among those calling for stronger laws to protect human musicians. At the same time, scammers have exploited AI music tools to fraudulently earn streaming revenue on platforms like Spotify.
Over the past year, Spotify have seen a surge of AI-generated content. Some virtual bands, like The Velvet Sundown, have successfully fooled listeners into thinking they're real, amassing millions of views on platforms such as YouTube. Earlier this year, a parody track with controversial lyrics, created using the AI tool Udio, even climbed to number three on Spotify's viral song charts.
Track Latest News Live on NDTV.com and get news updates from India and around the world