https://scholars.lib.ntu.edu.tw/handle/123456789/633676
標題: | Improving Generalizability of Distilled Self-Supervised Speech Processing Models Under Distorted Settings | 作者: | Huang, Kuan Po Fu, Yu Kuan Hsu, Tsu Yuan Gutierrez, Fabian Ritter Wang, Fan Lin Tseng, Liang Hsuan Zhang, Yu HUNG-YI LEE |
關鍵字: | Distortions | Domain Adversarial Training | Domain-adaptive Pre-training | Self-supervised Learning | SUPERB | 公開日期: | 1-一月-2023 | 來源出版物: | 2022 IEEE Spoken Language Technology Workshop, SLT 2022 - Proceedings | 摘要: | Self-supervised learned (SSL) speech pre-trained models perform well across various speech processing tasks. Distilled versions of SSL models have been developed to match the needs of on-device speech applications. Though having similar performance as original SSL models, distilled counterparts suffer from performance degradation even more than their original versions in distorted environments. This paper proposes to apply Cross-Distortion Mapping and Domain Adversarial Training to SSL models during knowledge distillation to alleviate the performance gap caused by the domain mismatch problem. Results show consistent performance improvements under both in- and out-of-domain distorted setups for different downstream tasks while keeping efficient model size. |
URI: | https://scholars.lib.ntu.edu.tw/handle/123456789/633676 | ISBN: | 9798350396904 | DOI: | 10.1109/SLT54892.2023.10022474 |
顯示於: | 電機工程學系 |
在 IR 系統中的文件,除了特別指名其著作權條款之外,均受到著作權保護,並且保留所有的權利。