Abstract
We tackle the task of conditional music generation. We introduce MUSICGEN, a single Language Model (LM) that operates over several streams of compressed discrete music representation, i.e., tokens. Unlike prior work, MUSICGEN is comprised of a single-stage transformer LM together with efficient token interleaving patterns, which eliminates the need for cascading several models, e.g., hierarchically or up-sampling. Following this approach, we demonstrate how MUSICGEN can generate high-quality samples, both mono and stereo, while being conditioned on textual description or melodic features, allowing better controls over the generated output. We conduct extensive empirical evaluation, considering both automatic and human studies, showing the proposed approach is superior to the evaluated baselines on a standard text-to-music benchmark. Through ablation studies, we shed light over the importance of each of the components comprising MUSICGEN. Music samples, code, and models are available at github.com/facebookresearch/audiocraft.
Original language | English |
---|---|
Pages (from-to) | 47704-47720 |
Number of pages | 17 |
Journal | Advances in Neural Information Processing Systems |
Volume | 36 |
State | Published - 2023 |
Externally published | Yes |
Event | 37th Conference on Neural Information Processing Systems, NeurIPS 2023 - New Orleans, United States Duration: 10 Dec 2023 → 16 Dec 2023 |
Bibliographical note
Publisher Copyright:© 2023 Neural information processing systems foundation. All rights reserved.