TDT-KWS: Fast And Accurate Keyword Spotting Using Token-and-duration Transducer
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2024
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866911804678995968 |
|---|---|
| author | Xi, Yu Li, Hao Yang, Baochen Li, Haoyu Xu, Hainan Yu, Kai |
| author_facet | Xi, Yu Li, Hao Yang, Baochen Li, Haoyu Xu, Hainan Yu, Kai |
| contents | Designing an efficient keyword spotting (KWS) system that delivers exceptional performance on resource-constrained edge devices has long been a subject of significant attention. Existing KWS search algorithms typically follow a frame-synchronous approach, where search decisions are made repeatedly at each frame despite the fact that most frames are keyword-irrelevant. In this paper, we propose TDT-KWS, which leverages token-and-duration Transducers (TDT) for KWS tasks. We also propose a novel KWS task-specific decoding algorithm for Transducer-based models, which supports highly effective frame-asynchronous keyword search in streaming speech scenarios. With evaluations conducted on both the public Hey Snips and self-constructed LibriKWS-20 datasets, our proposed KWS-decoding algorithm produces more accurate results than conventional ASR decoding algorithms. Additionally, TDT-KWS achieves on-par or better wake word detection performance than both RNN-T and traditional TDT-ASR systems while achieving significant inference speed-up. Furthermore, experiments show that TDT-KWS is more robust to noisy environments compared to RNN-T KWS. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2403_13332 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | TDT-KWS: Fast And Accurate Keyword Spotting Using Token-and-duration Transducer Xi, Yu Li, Hao Yang, Baochen Li, Haoyu Xu, Hainan Yu, Kai Audio and Speech Processing Sound Designing an efficient keyword spotting (KWS) system that delivers exceptional performance on resource-constrained edge devices has long been a subject of significant attention. Existing KWS search algorithms typically follow a frame-synchronous approach, where search decisions are made repeatedly at each frame despite the fact that most frames are keyword-irrelevant. In this paper, we propose TDT-KWS, which leverages token-and-duration Transducers (TDT) for KWS tasks. We also propose a novel KWS task-specific decoding algorithm for Transducer-based models, which supports highly effective frame-asynchronous keyword search in streaming speech scenarios. With evaluations conducted on both the public Hey Snips and self-constructed LibriKWS-20 datasets, our proposed KWS-decoding algorithm produces more accurate results than conventional ASR decoding algorithms. Additionally, TDT-KWS achieves on-par or better wake word detection performance than both RNN-T and traditional TDT-ASR systems while achieving significant inference speed-up. Furthermore, experiments show that TDT-KWS is more robust to noisy environments compared to RNN-T KWS. |
| title | TDT-KWS: Fast And Accurate Keyword Spotting Using Token-and-duration Transducer |
| topic | Audio and Speech Processing Sound |
| url | https://arxiv.org/abs/2403.13332 |