doi: 10.18178/ijfcc.2016.5.2.458
Investigation of DNN-Based Keyword Spotting in Low Resource Environments
Kaixiang Shen
,
Meng Cai
,
Wei-Qiang Zhang
,
Yao Tian
,
Jia Liu
- Department of Electronic Engineering, Tsinghua University, China.
Abstract
Keyword Spotting is a challenging task aiming atdetecting the predefined keywords in utterances. In the lowresource environment such as little keyword templates and thelack of linguistic information, the detection performance isalways unsatisfactory. In this paper, we focus on the lowresource situation where every keyword only has about 40templates and the linguistic information is unknown. We exploreusing deep neural networks for acoustic modeling. In addition,we investigate several techniques including transfer-learning,multilingual bottleneck features, balancing keyword filler dataand data augmentation to address the low resource problem andimprove the system's performance. Compared with aquery-by-example baseline system, substantial performanceimprovement can be obtained with our proposed keywordspotting system with deep neural network (KWS-DNN)framework.
Keywords
- Keyword spotting
- DNN
- acoustic model
How to Cite
Kaixiang Shen, Meng Cai, Wei-Qiang Zhang, Yao Tian, and Jia Liu, "Investigation of DNN-Based Keyword Spotting in Low Resource Environments," International Journal of Future Computer and Communication, vol. 5, no. 2, pp. 125-129, 2016. https://doi.org/ 10.18178/ijfcc.2016.5.2.458
Copyright & License
Copyright © 2016 by the authors. This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited (CC BY 4.0).