Saved in:
Bibliographic Details
Main Authors: Gao, Xiaoxue, Chen, Nancy F.
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2409.18654
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909327837626368
author Gao, Xiaoxue
Chen, Nancy F.
author_facet Gao, Xiaoxue
Chen, Nancy F.
contents Current automatic speech recognition systems struggle with modeling long speech sequences due to high quadratic complexity of Transformer-based models. Selective state space models such as Mamba has performed well on long-sequence modeling in natural language processing and computer vision tasks. However, research endeavors in speech technology tasks has been under-explored. We propose Speech-Mamba, which incorporates selective state space modeling in Transformer neural architectures. Long sequence representations with selective state space models in Speech-Mamba is complemented with lower-level representations from Transformer-based modeling. Speech-mamba achieves better capacity to model long-range dependencies, as it scales near-linearly with sequence length.
format Preprint
id arxiv_https___arxiv_org_abs_2409_18654
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Speech-Mamba: Long-Context Speech Recognition with Selective State Spaces Models
Gao, Xiaoxue
Chen, Nancy F.
Audio and Speech Processing
Sound
Current automatic speech recognition systems struggle with modeling long speech sequences due to high quadratic complexity of Transformer-based models. Selective state space models such as Mamba has performed well on long-sequence modeling in natural language processing and computer vision tasks. However, research endeavors in speech technology tasks has been under-explored. We propose Speech-Mamba, which incorporates selective state space modeling in Transformer neural architectures. Long sequence representations with selective state space models in Speech-Mamba is complemented with lower-level representations from Transformer-based modeling. Speech-mamba achieves better capacity to model long-range dependencies, as it scales near-linearly with sequence length.
title Speech-Mamba: Long-Context Speech Recognition with Selective State Spaces Models
topic Audio and Speech Processing
Sound
url https://arxiv.org/abs/2409.18654