Search results
30 apr. 2020 · Jukebox is a model that learns to produce raw audio music from lyrics, genre, and artist style. It uses a VQ-VAE to compress and decompress the audio, and transformers to generate the compressed codes.
Explore the samples of music generated by Jukebox, a model that can create novel and realistic songs in various genres and styles. Listen to re-renditions, continuations, unseen lyrics, and more from different artists and eras.
Jukebox: A Generative Model for Audio.
Jukebox is a code repository for a paper that presents a generative model for music. It includes instructions for sampling, training, and prompting the model with your own music.
Jukebox is a model that generates music with singing in the raw audio domain using a hierarchical VQ-VAE and autoregressive Transformers. It can produce diverse and coherent songs of various genres, styles, and lyrics, and is released by OpenAI.
30 apr. 2020 · Jukebox is a model that generates music with singing in the raw audio domain. It uses a multi-scale VQ-VAE and autoregressive Transformers to compress and model the long context of raw audio, and can condition on artist, genre and lyrics.
!pip install git+https://github.com/openai/jukebox.git IMPORTANT NOTE ON SYSTEM REQUIREMENTS: If you are connecting to a hosted runtime, make sure it has a P100 GPU...
