OpenAI is developing a new tool that allows creating music based on text and audio requests. Sources report that users will be able to add music to existing videos or create guitar accompaniment for vocal tracks. It is currently unknown when the company plans to launch this product and whether it will be a standalone service or integrated into ChatGPT or the video app Sora.
For preparing training data, OpenAI collaborates with several students from Juilliard School who help annotate notes. Previously, the company had already launched generative music models, but they appeared before the launch of ChatGPT. Recently, OpenAI has been working on audio models that convert text to speech and vice versa.
Similar tools for creating music already exist at Google and Suno. However, the new OpenAI product could become a convenient solution for those who want to quickly add music to videos or create musical accompaniment for vocals. Users will be able to use simple text or audio requests to get the desired result.
OpenAI has not yet commented on the details and timing of the new tool’s launch. The service could be useful for video creators, musicians, and anyone creating content using AI.
