Application of latent semantic analysis in continuous speech recognition

doi:10.3778/j.issn.1002-8331.2009.32.035

Computer Engineering and Applications ›› 2009, Vol. 45 ›› Issue (32): 111-113.DOI: 10.3778/j.issn.1002-8331.2009.32.035

• 数据库、信号与信息处理 • Previous Articles Next Articles

Application of latent semantic analysis in continuous speech recognition

OU Jian-lin，LIN Qian，SHI Xiao-dong

Department of Computer Science，Xiamen University，Xiamen，Fujian 361005，China

Received:2008-12-04 Revised:2009-02-19 Online:2009-11-11 Published:2009-11-11
Contact: OU Jian-lin

潜在语义分析在连续语音识别中的应用

欧建林，林茜，史晓东

厦门大学计算机科学系，福建厦门 361005

通讯作者: 欧建林

Abstract

Abstract: The theory of Latent Semantic Analysis（LSA） for speech recognition is described，and the related techniques for implementing LSA-based language modeling in speech recognition systems are presented.An LSA-based semantic model is constructed on the WSJ0 text corpus.This paper uses the interpolation method to combine this semantic model with conventional 3-gram to form a hybrid language model（i.e.，LSA+3-gram）.To optimize the performance of the hybrid model，it applies k-means algorithm to perform vector clustering in the LSA vector space while the density function is used to initialize the centroid.The constructed hybrid language model outperforms the corresponding 3-gram baseline：Continuous speech recognition experiments conducted on the WSJ0 test corpus show a relative reduction in word error rate of about 13.3%.

Key words: latent semantic analysis, N-gram, k-means clustering, continuous speech recognition

摘要： 研究了潜在语义分析（LSA）理论及其在连续语音识别中应用的相关技术，在此基础上利用WSJ0文本语料库上构建LSA模型，并将其与3-gram模型进行插值组合，构建了包含语义信息的统计语言模型；同时为了进一步优化混合模型的性能，提出了基于密度函数初始化质心的k-means聚类算法对LSA模型的向量空间进行聚类。WSJ0语料库上的连续语音识别实验结果表明：LSA+3-gram混合模型能够使识别的词错误率相比较于标准的3-gram下降13.3%。

关键词: 潜在语义分析, N元文法, k均值聚类, 连续语音识别

CLC Number:

TP391

OU Jian-lin，LIN Qian，SHI Xiao-dong. Application of latent semantic analysis in continuous speech recognition[J]. Computer Engineering and Applications, 2009, 45(32): 111-113.

欧建林，林茜，史晓东. 潜在语义分析在连续语音识别中的应用[J]. 计算机工程与应用, 2009, 45(32): 111-113.

[1]	WANG Changlong, ZHANG Yuandong, MIAO Hong, YANG Yuheng. Application of Double Channel Convolutional Neural Network in Pumpkin Diseases Identification [J]. Computer Engineering and Applications, 2021, 57(5): 183-189.
[2]	ZHANG Ziran, HUANG Weihua, CHEN Yang, ZHANG Zheng, LI Ziyuan. Improved Ant Colony Path Planning Algorithm Based on Bidirectional Search [J]. Computer Engineering and Applications, 2021, 57(21): 270-277.
[3]	LU Junjie, HUANG Jinquan, LU Feng. Likelihood K-means Clustering for Gas Path Failure Diagnostics of Turbofan Engine [J]. Computer Engineering and Applications, 2020, 56(9): 136-141.
[4]	WANG Weihong, ZENG Yingjie. Collaborative Filtering Recommendation Algorithm Based on Clustering and User Preference [J]. Computer Engineering and Applications, 2020, 56(3): 68-73.
[5]	MA Jinghui, PAN Wei, WANG Ru. 3D Point Cloud Classification Based on K-means Clustering [J]. Computer Engineering and Applications, 2020, 56(17): 181-186.
[6]	GUO Yongkun, ZHANG Xinyou, LIU Liping, DING Liang, NIU Xiaolu. K-means Clustering Algorithm of Optimizing Initial Clustering Center [J]. Computer Engineering and Applications, 2020, 56(15): 172-178.
[7]	LIU Qiang1, SHI Hong1, WANG Pingxin2，3, YANG Xibei1. Three-Way Clustering Analysis Based on [ε] Neighborhood [J]. Computer Engineering and Applications, 2019, 55(6): 140-144.
[8]	XI Runping1，2, JIA Gaoyun1，2, ZHANG Yanning1，2, ZHANG Fujun1，2. Objects Detection in Different Source Images Based on Evaluation Vector [J]. Computer Engineering and Applications, 2019, 55(1): 180-185.
[9]	DONG Benzhi, NIE Lili, JING Weipeng, CUI Hang. Identification method of ambrostoma quadriimpressum motschlsky based on Faster R-CNN [J]. Computer Engineering and Applications, 2018, 54(23): 89-93.
[10]	HAN Chong1, YUAN Yingshan2, MEI Tao2, GENG Huiling2. Data stream outlier detection algorithm based on K-means [J]. Computer Engineering and Applications, 2017, 53(3): 58-63.
[11]	SUN Yuexin, MA Huifang, YAO Wei, ZHANG Zhichang. Microblog hot topic detection based on positive point mutual information and probabilistic topic model [J]. Computer Engineering and Applications, 2016, 52(6): 61-66.
[12]	ZHANG Jinmin1, LEI Jiang1，2. Background subtraction algorithm using statistical modeling and object updating mechanism [J]. Computer Engineering and Applications, 2016, 52(11): 152-157.
[13]	WANG Qian, LIU Shenghui. Application research of improved K-means algorithm in intrusion detection [J]. Computer Engineering and Applications, 2015, 51(17): 124-127.
[14]	WANG Xiuzhen, CONG Rui, WANG Fei. Novel spelling correction algorithm for online query [J]. Computer Engineering and Applications, 2015, 51(14): 113-119.
[15]	DUAN Baobin1，2, HAN Lixin2. Improved convolutional neural networks and its application in stitching of scrapped paper [J]. Computer Engineering and Applications, 2014, 50(9): 176-181.

Application of latent semantic analysis in continuous speech recognition

潜在语义分析在连续语音识别中的应用

PDF

Knowledge

Abstract

Cite this article

share this article

References

Related Articles 15

Recommended Articles

Metrics