Jukebox: A Generative Model for Music
This addresses the problem of controllable music generation for creators and AI researchers, representing a novel method for a known bottleneck.
The researchers tackled generating music with singing in raw audio by using a multi-scale VQ-VAE and autoregressive Transformers, resulting in high-fidelity and diverse songs with coherence up to multiple minutes, as demonstrated by thousands of released samples.
We introduce Jukebox, a model that generates music with singing in the raw audio domain. We tackle the long context of raw audio using a multi-scale VQ-VAE to compress it to discrete codes, and modeling those using autoregressive Transformers. We show that the combined model at scale can generate high-fidelity and diverse songs with coherence up to multiple minutes. We can condition on artist and genre to steer the musical and vocal style, and on unaligned lyrics to make the singing more controllable. We are releasing thousands of non cherry-picked samples at https://jukebox.openai.com, along with model weights and code at https://github.com/openai/jukebox