Using output probability distribution for improving speech recognition in adverse environment

Shilei Huang; Xiang Xie; Jingming Kuang

Using output probability distribution for improving speech recognition in adverse environment

Shilei Huang^*
, Xiang Xie
, Jingming Kuang

^*此作品的通讯作者

信息与电子学院

Beijing Institute of Technology

科研成果: 会议稿件 › 论文 › 同行评审

摘要

This paper proposed a method to improve the accuracy of small vocabulary isolated word speaker-independent speech recognition in adverse environment. The proposed approach is implemented by using Output Probability Distributions (OPDs) and Support Vector Machine (SVM). OPDs improve the system performance by modeling inter-word relationships; then SVM classifiers are used to discriminate the difference between OPD models. The system was tested using isolated Mandarin digits database, corrupted with the NOISEX-92 database. The experiments have achieved good result in noise conditions, the WER dropped about 30% on average when compared to the HMM recognizer.

源语言	英语
页	929-932
页数	4
出版状态	已出版 - 2005
活动	9th European Conference on Speech Communication and Technology - Lisbon, 葡萄牙期限: 4 9月 2005 → 8 9月 2005

会议

会议	9th European Conference on Speech Communication and Technology
国家/地区	葡萄牙
市	Lisbon
时期	4/09/05 → 8/09/05

其它文件与链接

链接到 Scopus 的出版物

引用此

@conference{61c4e70af9a74ee597025828c93b0bf1,

title = "Using output probability distribution for improving speech recognition in adverse environment",

abstract = "This paper proposed a method to improve the accuracy of small vocabulary isolated word speaker-independent speech recognition in adverse environment. The proposed approach is implemented by using Output Probability Distributions (OPDs) and Support Vector Machine (SVM). OPDs improve the system performance by modeling inter-word relationships; then SVM classifiers are used to discriminate the difference between OPD models. The system was tested using isolated Mandarin digits database, corrupted with the NOISEX-92 database. The experiments have achieved good result in noise conditions, the WER dropped about 30\% on average when compared to the HMM recognizer.",

author = "Shilei Huang and Xiang Xie and Jingming Kuang",

year = "2005",

language = "English",

pages = "929--932",

note = "9th European Conference on Speech Communication and Technology ; Conference date: 04-09-2005 Through 08-09-2005",

}

TY - CONF

T1 - Using output probability distribution for improving speech recognition in adverse environment

AU - Huang, Shilei

AU - Xie, Xiang

AU - Kuang, Jingming

PY - 2005

Y1 - 2005

N2 - This paper proposed a method to improve the accuracy of small vocabulary isolated word speaker-independent speech recognition in adverse environment. The proposed approach is implemented by using Output Probability Distributions (OPDs) and Support Vector Machine (SVM). OPDs improve the system performance by modeling inter-word relationships; then SVM classifiers are used to discriminate the difference between OPD models. The system was tested using isolated Mandarin digits database, corrupted with the NOISEX-92 database. The experiments have achieved good result in noise conditions, the WER dropped about 30% on average when compared to the HMM recognizer.

AB - This paper proposed a method to improve the accuracy of small vocabulary isolated word speaker-independent speech recognition in adverse environment. The proposed approach is implemented by using Output Probability Distributions (OPDs) and Support Vector Machine (SVM). OPDs improve the system performance by modeling inter-word relationships; then SVM classifiers are used to discriminate the difference between OPD models. The system was tested using isolated Mandarin digits database, corrupted with the NOISEX-92 database. The experiments have achieved good result in noise conditions, the WER dropped about 30% on average when compared to the HMM recognizer.

UR - https://www.scopus.com/pages/publications/33745192530

M3 - Paper

AN - SCOPUS:33745192530

SP - 929

EP - 932

T2 - 9th European Conference on Speech Communication and Technology

Y2 - 4 September 2005 through 8 September 2005

ER -

Using output probability distribution for improving speech recognition in adverse environment

摘要

会议

其它文件与链接

指纹

引用此