Audio Mamba: Bidirectional State Space Model for Audio Representation Learning
IEEE Signal Processing Letters · 2024 · 31 · 2975-2979
DOI: 10.1109/lsp.2024.3483009 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Signal Processing Letters, 2024. Verfasst von Mehmet Hamza Erol und 3 weiteren Autor:innen. Die Arbeit wurde bisher 55-mal zitiert (Crossref) und nennt 36 Literaturangaben. Gefördert von: National Research Foundation of Korea, Korean Government, Information Technology Research Center.
Abstract (Originalsprache)
Transformers have rapidly become the preferred choice for audio classification, surpassing methods based on CNNs. However, Audio Spectrogram Transformers (ASTs) exhibit quadratic scaling due to self-attention. The removal of this quadratic self-attention cost presents an appealing direction. Recently, state space models (SSMs), such as Mamba, have demonstrated potential in language and vision tasks in this regard. In this study, we explore whether reliance on self-attention is necessary for audio classification tasks. By introducing Audio Mamba (AuM), the first self-attention-free, purely SSM-based model for audio classification, we aim to address this question. We evaluate AuM on various audio datasets - comprising six different benchmarks - where it achieves comparable or better performance compared to well-established AST model.
Förderung
- National Research Foundation of Korea
- Korean Government
- Information Technology Research Center
- Institute for Information and Communications Technology Planning and Evaluation
Zitieren
APA
Erol, M. H., Senocak, A., Feng, J., & Chung, J. S. (2024). Audio Mamba: Bidirectional State Space Model for Audio Representation Learning. IEEE Signal Processing Letters, 31, 2975-2979. https://doi.org/10.1109/lsp.2024.3483009
Vancouver
Erol MH, Senocak A, Feng J, Chung JS. Audio Mamba: Bidirectional State Space Model for Audio Representation Learning. IEEE Signal Processing Letters. 2024;31:2975-2979. doi:10.1109/lsp.2024.3483009
BibTeX
@article{Erol2024,
author = {Erol, Mehmet Hamza and Senocak, Arda and Feng, Jiu and Chung, Joon Son},
title = {Audio Mamba: Bidirectional State Space Model for Audio Representation Learning},
journal = {IEEE Signal Processing Letters},
year = {2024},
volume = {31},
pages = {2975--2979},
doi = {10.1109/lsp.2024.3483009}
}
EUVIMED