OpenAI developing new generative music tool capable of creating compositions from text and audio prompts
Although OpenAI has previously developed generative music models, those efforts predate the introduction of ChatGPT.
ADVERTISEMENT
OpenAI is reportedly developing a new artificial intelligence tool designed to generate music from text and audio prompts, according to a report by The Information.
As per a report by TechCrunch, sources familiar with the project said the upcoming model could be used to create original soundtracks for videos or even generate instrumental accompaniments, such as guitar backing for an existing vocal track. It remains unclear when the company intends to launch the tool or whether it will be released as a standalone product or integrated into OpenAI’s existing platforms, such as ChatGPT or the video-generation tool Sora.
One source told The Information that OpenAI is collaborating with students from the Juilliard School, one of the world’s leading music conservatories, to annotate musical scores. This annotated data will reportedly be used to help train the model to understand and replicate musical structure, tone, and rhythm more accurately.
Although OpenAI has previously developed generative music models, those efforts predate the introduction of ChatGPT. In recent years, the company’s focus has shifted more towards audio systems that handle text-to-speech and speech-to-text functionalities.
The move would place OpenAI in direct competition with other players in the generative music space, including Google, which has been advancing its MusicLM technology, and Suno, another AI startup specialising in music generation.
If successful, OpenAI’s latest project could mark a significant expansion of its creative AI portfolio, bridging text, video, and music in a single ecosystem.
