Methods › Audio › Generative Audio Models › Jukebox
Jukebox
archive 2025-07-28 Description, source and code snippet are the archive's method entry.
Jukebox is a model that generates music with singing in the raw audio domain. It tackles the long context of raw audio using a multi-scale VQ-VAE to compress it to discrete codes, and modeling those using autoregressive Transformers. It can condition on artist and genre to steer the musical and vocal style, and on unaligned lyrics to make the singing more controllable.
Three separate VQ-VAE models are trained with different temporal resolutions. At each level, the input audio is segmented and encoded into latent vectors 𝐡ₜ, which are then quantized to the closest codebook vectors 𝐞_(zₜ). The code zₜ is a discrete representation of the audio that we later train our prior on. The decoder takes the sequence of codebook vectors and reconstructs the audio. The top level learns the highest degree of abstraction, since it is encoding longer audio per token while keeping the codebook size the same. Audio can be reconstructed using the codes at any one of the abstraction levels, where the least abstract bottom-level codes result in the highest-quality audio.
Papers archive 2025-07-28
13 shown of 13, newest first. Repository counts are the archive's code-links table. A Syntology line states what Syntology ran from that paper's harvested code; it is per sample and not a correctness claim.
-
Comparative Analysis of Pretrained Audio Representations in Music Recommender Systems 13 Sep 2024 · 1 repository · arXiv:2409.08987
-
An End-to-End Approach for Chord-Conditioned Song Generation 10 Sep 2024 · 0 repositories · arXiv:2409.06307
-
From Audio Encoders to Piano Judges: Benchmarking Performance Understanding for Solo Piano 5 Jul 2024 · 0 repositories · arXiv:2407.04518
-
A Novel Audio Representation for Music Genre Identification in MIR 1 Apr 2024 · 0 repositories · arXiv:2404.01058
-
Advancements in Generative AI: A Comprehensive Review of GANs, GPT, Autoencoders, Diffusion Model, and Transformers 17 Nov 2023 · 0 repositories · arXiv:2311.10242
-
MAP-Music2Vec: A Simple and Effective Baseline for Self-Supervised Music Audio Representation Learning 5 Dec 2022 · 0 repositories · arXiv:2212.02508
-
Melody transcription via generative pre-training 4 Dec 2022 · 1 repository · arXiv:2212.01884
-
EDGE: Editable Dance Generation From Music 19 Nov 2022 · 1 repository · arXiv:2211.10658Syntology ran 7 of 11 samples · 4 unverified
-
Evaluating Deep Music Generation Methods Using Data Augmentation 31 Dec 2021 · 0 repositories · arXiv:2201.00052
-
Transfer Learning with Jukebox for Music Source Separation 28 Nov 2021 · 1 repository · arXiv:2111.14200
-
Unsupervised Source Separation By Steering Pretrained Music Models 25 Oct 2021 · 1 repository · arXiv:2110.13071
-
Codified audio language modeling learns useful representations for music information retrieval 12 Jul 2021 · 1 repository · arXiv:2107.05677Syntology ran 3 of 3 samples · 0 unverified
-
Jukebox: A Generative Model for Music 30 Apr 2020 · 12 repositories · arXiv:2005.00341
Tasks archive 2025-07-28
20 shown of 30 tasks the archive attaches to papers tagged with this method, by distinct papers. A task without a page in the catalog is plain text.
Usage over time archive 2025-07-28
Components: the archive holds no method-to-method composition, so PwC's Components table cannot be rebuilt; the Papers list carries no Results column for the same reason (the archive does not join its leaderboard rows to method tags).
Categories archive 2025-07-28
Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections