OpenAI is developing a new tool that can generate music from text and audio prompts, according to a report from The Information. This technology could be used to add background music to videos or to provide a guitar accompaniment for a vocal track. It remains unclear when OpenAI intends to launch this tool or if it will be a standalone product instead of being integrated into existing services like ChatGPT or the video app Sora.
Sources indicate that OpenAI is collaborating with students from the Juilliard School to annotate musical scores. This work is intended to create training data for the artificial intelligence. While OpenAI has created generative music models in the past, those efforts came before the launch of ChatGPT. The company’s more recent audio work has concentrated on text-to-speech and speech-to-text models. Other companies, including Google and Suno, are also known to be working on their own generative music models. TechCrunch has contacted OpenAI for a statement.