Optimizing Dysarthria Wake-Up Word Spotting: An End-to-End Approach for SLT 2024 LRDWWS Challenge
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866916396111233024 |
|---|---|
| author | Liu, Shuiyun Kong, Yuxiang Guo, Pengcheng Zhuang, Weiji Gao, Peng Wang, Yujun Xie, Lei |
| author_facet | Liu, Shuiyun Kong, Yuxiang Guo, Pengcheng Zhuang, Weiji Gao, Peng Wang, Yujun Xie, Lei |
| contents | Speech has emerged as a widely embraced user interface across diverse applications. However, for individuals with dysarthria, the inherent variability in their speech poses significant challenges. This paper presents an end-to-end Pretrain-based Dual-filter Dysarthria Wake-up word Spotting (PD-DWS) system for the SLT 2024 Low-Resource Dysarthria Wake-Up Word Spotting Challenge. Specifically, our system improves performance from two key perspectives: audio modeling and dual-filter strategy. For audio modeling, we propose an innovative 2branch-d2v2 model based on the pre-trained data2vec2 (d2v2), which can simultaneously model automatic speech recognition (ASR) and wake-up word spotting (WWS) tasks through a unified multi-task finetuning paradigm. Additionally, a dual-filter strategy is introduced to reduce the false accept rate (FAR) while maintaining the same false reject rate (FRR). Experimental results demonstrate that our PD-DWS system achieves an FAR of 0.00321 and an FRR of 0.005, with a total score of 0.00821 on the test-B eval set, securing first place in the challenge. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2409_10076 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Optimizing Dysarthria Wake-Up Word Spotting: An End-to-End Approach for SLT 2024 LRDWWS Challenge Liu, Shuiyun Kong, Yuxiang Guo, Pengcheng Zhuang, Weiji Gao, Peng Wang, Yujun Xie, Lei Sound Human-Computer Interaction Audio and Speech Processing Speech has emerged as a widely embraced user interface across diverse applications. However, for individuals with dysarthria, the inherent variability in their speech poses significant challenges. This paper presents an end-to-end Pretrain-based Dual-filter Dysarthria Wake-up word Spotting (PD-DWS) system for the SLT 2024 Low-Resource Dysarthria Wake-Up Word Spotting Challenge. Specifically, our system improves performance from two key perspectives: audio modeling and dual-filter strategy. For audio modeling, we propose an innovative 2branch-d2v2 model based on the pre-trained data2vec2 (d2v2), which can simultaneously model automatic speech recognition (ASR) and wake-up word spotting (WWS) tasks through a unified multi-task finetuning paradigm. Additionally, a dual-filter strategy is introduced to reduce the false accept rate (FAR) while maintaining the same false reject rate (FRR). Experimental results demonstrate that our PD-DWS system achieves an FAR of 0.00321 and an FRR of 0.005, with a total score of 0.00821 on the test-B eval set, securing first place in the challenge. |
| title | Optimizing Dysarthria Wake-Up Word Spotting: An End-to-End Approach for SLT 2024 LRDWWS Challenge |
| topic | Sound Human-Computer Interaction Audio and Speech Processing |
| url | https://arxiv.org/abs/2409.10076 |