Small-Footprint Magic Word Detection Method Using Convolutional LSTM Neural Network

被引:6
|
作者
Yamamoto, Taiki [1 ]
Nishimura, Ryota [1 ]
Misaki, Masayuki [2 ]
Kitaoka, Norihide [1 ]
机构
[1] Tokushima Univ, Dept Adv Technol & Sci, Tokushima, Japan
[2] Panasonic Corp, Osaka, Japan
来源
关键词
keyword spotting; convolutional neural network; recurrent neural network; convolutional LSTM; small footprint;
D O I
10.21437/Interspeech.2019-1662
中图分类号
R36 [病理学]; R76 [耳鼻咽喉科学];
学科分类号
100104 ; 100213 ;
摘要
The number of consumer devices which can be operated by voice is increasing every year. Magic Word Detection (MWD), the detection of an activation keyword in continuous speech, has become an essential technology for the hands-free operation of such devices. Because MWD systems need to run constantly in order to detect Magic Words at any time, many studies have focused on the development of a small-footprint system. In this paper, we propose a novel, small-footprintMWDmethod which uses a convolutional Long Short-Term Memory (LSTM) neural network to capture frequency and time domain features over time. As a result, the proposed method outperforms the baseline method while reducing the number of parameters by more than 80%. An experiment on a small-scale device demonstrates that our model is efficient enough to function in real time.
引用
收藏
页码:2035 / 2039
页数:5
相关论文
共 50 条
  • [21] Small-Footprint Keyword Spotting Based on Gated Channel Transformation Sandglass Residual Neural Network
    Zhang, Ying
    Zhu, Shirong
    Yu, Chao
    Zhao, Lasheng
    INTERNATIONAL JOURNAL OF PATTERN RECOGNITION AND ARTIFICIAL INTELLIGENCE, 2022, 36 (07)
  • [22] Region Proposal Network Based Small-Footprint Keyword Spotting
    Hou, Jingyong
    Shi, Yangyang
    Ostendorf, Mari
    Hwang, Mei-Yuh
    Xie, Lei
    IEEE SIGNAL PROCESSING LETTERS, 2019, 26 (10) : 1471 - 1475
  • [23] Android malware classification using convolutional neural network and LSTM
    Hosseini, Soodeh
    Nezhad, Ali Emamali
    Seilani, Hossein
    JOURNAL OF COMPUTER VIROLOGY AND HACKING TECHNIQUES, 2021, 17 (04) : 307 - 318
  • [24] Dysarthric Speech Recognition Using Convolutional LSTM Neural Network
    Kim, Myungjong
    Cao, Beiming
    An, Kwanghoon
    Wang, Jun
    19TH ANNUAL CONFERENCE OF THE INTERNATIONAL SPEECH COMMUNICATION ASSOCIATION (INTERSPEECH 2018), VOLS 1-6: SPEECH RESEARCH FOR EMERGING MARKETS IN MULTILINGUAL SOCIETIES, 2018, : 2948 - 2952
  • [25] Android malware classification using convolutional neural network and LSTM
    Soodeh Hosseini
    Ali Emamali Nezhad
    Hossein Seilani
    Journal of Computer Virology and Hacking Techniques, 2021, 17 : 307 - 318
  • [26] A Convolutional Neural Network-Based Method for Small Traffic Sign Detection
    Zhou S.
    Zhi X.
    Liu D.
    Ning H.
    Jiang L.
    Shi F.
    Tongji Daxue Xuebao/Journal of Tongji University, 2019, 47 (11): : 1626 - 1632
  • [27] Method of Profanity Detection Using Word Embedding and LSTM
    Yi, MoungHo
    Lim, MyungJin
    Ko, Hoon
    Shin, JuHyun
    MOBILE INFORMATION SYSTEMS, 2021, 2021
  • [28] A Time Delay Neural Network with Shared Weight Self-Attention for Small-Footprint Keyword Spotting
    Bai, Ye
    Yi, Jiangyan
    Tao, Jianhua
    Wen, Zhengqi
    Tian, Zhengkun
    Zhao, Chenghao
    Fan, Cunhang
    INTERSPEECH 2019, 2019, : 2190 - 2194
  • [29] A Grasp Detection Method for Industrial Robots Using a Convolutional Neural Network
    Ogas, E.
    Avila, L.
    Larregay, G.
    Moran, D.
    IEEE LATIN AMERICA TRANSACTIONS, 2019, 17 (09) : 1509 - 1516
  • [30] A Robust Abnormal Behavior Detection Method Using Convolutional Neural Network
    Tay, Nian Chi
    Connie, Tee
    Ong, Thian Song
    Goh, Kah Ong Michael
    Teh, Pin Shen
    COMPUTATIONAL SCIENCE AND TECHNOLOGY, 2019, 481 : 37 - 47