OpenAI is reportedly developing a new tool that would generate music based on text and audio prompts, as revealed by a report from The Information. This tool is envisioned to enable users to seamlessly incorporate music into videos or even add guitar accompaniment to existing vocal tracks.
The specific launch timeline for this tool remains undisclosed, and it is uncertain whether it will be a standalone product or integrated with OpenAI’s existing ChatGPT and video app, Sora. Notably, OpenAI is collaborating with students from the Juilliard School to annotate scores, enhancing the quality of training data for this music generation tool.
While OpenAI has previously introduced generative music models, these earlier models were introduced before ChatGPT. Lately, the company has been concentrating on developing audio models primarily focused on text-to-speech and speech-to-text functionalities. Other notable companies like Google and Suno are also actively exploring generative music models in this space.
TechCrunch has initiated contact with OpenAI for additional insights on this development.
Source: TechCrunch