Jukebox: A Generative Model for Music (AI Paper Summary)
OpenAI releases Jukebox, a machine learning framework that generates music
Paper: https://arxiv.org/pdf/2005.00341.pdf
Summary by: Luca Arrotta
Machine Learning Researcher (Italy)
OpenAI launched Jukebox, a model that generates music with singing in the raw audio domain. As a generative model for music, Jukebox can handle the long context of raw audio using an autoencoder. Jukebox’s autoencoder processes the audio files using a multiscale VQ-VAE to compress it to discrete codes and modeling those using autoregressive Transformers.
Provided with a genre, artist, and lyrics as input, Jukebox can output a new music sample produced from scratch. This is a type of innovation that expands the boundaries of generative models to a new level. Jukebox’s model is capable of generating audio pieces that are multiple minutes long, and with recognizable singing in natural-sounding voices. Please listen to the Jukebox-generated country song listed at the end of this article.
OpenAI releases Jukebox, a machine learning framework that generates music
Paper: https://arxiv.org/pdf/2005.00341.pdf
Summary by: Luca Arrotta
Machine Learning Researcher (Italy)
OpenAI launched Jukebox, a model that generates music with singing in the raw audio domain. As a generative model for music, Jukebox can handle the long context of raw audio using an autoencoder. Jukebox’s autoencoder processes the audio files using a multiscale VQ-VAE to compress it to discrete codes and modeling those using autoregressive Transformers.
Provided with a genre, artist, and lyrics as input, Jukebox can output a new music sample produced from scratch. This is a type of innovation that expands the boundaries of generative models to a new level. Jukebox’s model is capable of generating audio pieces that are multiple minutes long, and with recognizable singing in natural-sounding voices. Please listen to the Jukebox-generated country song listed at the end of this article.
