OpenAI tunes generative music

OpenAI is reportedly developing a new tool capable of generating music from both text and audio prompts. The unreleased tool could allow users to create custom soundtracks and musical compositions using AI. The model will likely leverage a deep learning architecture, trained on a vast dataset of music, to capture complex patterns in melody, harmony, rhythm, and timbre.

Such a tool could have a significant impact on the music industry, potentially democratising music creation and making bespoke soundtracks accessible to a wider audience. AI-generated music could offer solutions for content creators needing custom musical scores. Competitors to the unreleased music tool include AudioCraft and Suno.

OpenAI previously released Jukebox in 2020, a neural network that generates music, including rudimentary singing, as raw audio in a variety of genres and artist styles. Also, MuseNet was released in 2019, a deep neural network that can generate 4-minute musical compositions with 10 different instruments, and can combine styles from country to Mozart to the Beatles.