Improved features using convolution-augmented transformers for keyword spotting

ITM Web of Conferences

Yang Junan¹  Chen Qiang²  Li Song²  Liu Jingtao²  Wang Yi² 

[1] College of Electronic Countermeasures, National University of Defence Technology;Unit 91977 of PLA;

关键词: keyword spotting;    attention;    convolutional neural networks;    transformers;   

DOI  :  10.1051/itmconf/20224702039

来源: DOAJ

【 摘 要 】

Transformer can effectively model long rang dependency, but suffer from uncapable to extract local feature patterns. While CNNs exploit local features effectively. In this paper, we seek to combine convolution and Transformers improves over using them individually, and propose improved features using convolution-augmented transformers for keyword spotting. The convolution-augmented transformers are constructed with a ResNet front-end and a convolution-augmented transformers back-end in series. Using this improved feature for keyword spotting task. The results show that the improved features using convolution- augmented transformers can yield at least 3% improvement compared with other features.

【 授权许可】

Unknown