Initialization of Adaptation by Sufficient Statistics Using Phonetic Tree

摘要：

　　In this work we deal with the problem of small amount of data when estimating a feature transformation for the speaker adaptation of an acoustic model.Our goal is to compensate for the lack of adaptation data by a proper initialization of transformation matrices.Methods used in such situations are described,they are based on collecting additional accumulated statistics from nearest speakers.The proposed initialization approach is based on accumulated statistics too,but it incorporates also phonetic information when selecting the nearest statistics.Initialization methods compensating for the absence of actual speakers data are tested on telephone recordings with different amounts of adaptation data.In worst situation with extremely small amount of adaptation data relative improvement of 5% is obtained.

关键词： speech recognition adaptation initialization phonetic tree

作者: Zbyněk Zaji(c) Luká(s) Machlica Luděk Müiller

作者单位: Department of Cybernetics, Faculty of Applied Sciences University of West Bohemia, Plze(n), Czech Republic

会议类型: 国际会议

会议名称: 2012 IEEE 11th International Conference on Signal Processing (第11届IEEE信号处理国际会议)

会议地点: 北京

会议语种:英文

页码: 503-506

在线出版日期: 2012-10-21（万方平台首次上网日期，不代表论文的发表时间）