Work

Large-Scale Pretrained Model for Self-Supervised Music Audio Representation Learning

Year: 2022

Type: article

Abstract: Self-supervised learning technique is an under-explored topic for music audio due to the challenge of designing an appropriate training paradigm. We hence propose MAP-MERT, a large-scale music audio pre-trained model for general music understanding. We achieve performance that is comparable to the state-of-the-art pre-trained model Jukebox using less than 2% of parameters.

Source: Research Explorer (The University of Manchester)

Authors: Li, Y, Yuan, R, Zhang, G, Ma, Y, Lin, C, Chen, X, Ragni, A, Yin, H, Hu, Z, He, H, Benetos, E, Gyenge, N, Liu, R, Fu, J, DMRN+17: Digital Music Research Network One-day Workshop 2022

Institutions: University of Sheffield, Beijing Academy of Artificial Intelligence, Carnegie Mellon University, Queen Mary University of London, University of Michigan, Peking University, Max Planck Institute for Intelligent Systems, University of Tübingen, Dartmouth College

Language: English

Topic: Music and Audio Processing

Subfield: Signal Processing

Field: Computer Science

Domain: Physical Sciences

Open Access status: green

OpenAlex ID: W7011068324

FWCI:

Cites:

Cited by:

Related to: 10