Dynamic Gated Recurrent Neural Network for Compute-efficient Speech Enhancement
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914955591155712 |
|---|---|
| author | Cheng, Longbiao Pandey, Ashutosh Xu, Buye Delbruck, Tobi Liu, Shih-Chii |
| author_facet | Cheng, Longbiao Pandey, Ashutosh Xu, Buye Delbruck, Tobi Liu, Shih-Chii |
| contents | This paper introduces a new Dynamic Gated Recurrent Neural Network (DG-RNN) for compute-efficient speech enhancement models running on resource-constrained hardware platforms. It leverages the slow evolution characteristic of RNN hidden states over steps, and updates only a selected set of neurons at each step by adding a newly proposed select gate to the RNN model. This select gate allows the computation cost of the conventional RNN to be reduced during network inference. As a realization of the DG-RNN, we further propose the Dynamic Gated Recurrent Unit (D-GRU) which does not require additional parameters. Test results obtained from several state-of-the-art compute-efficient RNN-based speech enhancement architectures using the DNS challenge dataset, show that the D-GRU based model variants maintain similar speech intelligibility and quality metrics comparable to the baseline GRU based models even with an average 50% reduction in GRU computes. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2408_12425 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Dynamic Gated Recurrent Neural Network for Compute-efficient Speech Enhancement Cheng, Longbiao Pandey, Ashutosh Xu, Buye Delbruck, Tobi Liu, Shih-Chii Audio and Speech Processing Machine Learning Sound This paper introduces a new Dynamic Gated Recurrent Neural Network (DG-RNN) for compute-efficient speech enhancement models running on resource-constrained hardware platforms. It leverages the slow evolution characteristic of RNN hidden states over steps, and updates only a selected set of neurons at each step by adding a newly proposed select gate to the RNN model. This select gate allows the computation cost of the conventional RNN to be reduced during network inference. As a realization of the DG-RNN, we further propose the Dynamic Gated Recurrent Unit (D-GRU) which does not require additional parameters. Test results obtained from several state-of-the-art compute-efficient RNN-based speech enhancement architectures using the DNS challenge dataset, show that the D-GRU based model variants maintain similar speech intelligibility and quality metrics comparable to the baseline GRU based models even with an average 50% reduction in GRU computes. |
| title | Dynamic Gated Recurrent Neural Network for Compute-efficient Speech Enhancement |
| topic | Audio and Speech Processing Machine Learning Sound |
| url | https://arxiv.org/abs/2408.12425 |