Direct Construction of Compact Context-Dependency Transducers From Data

被引:0
作者
Rybach, David [1 ]
Riley, Michael [2 ]
机构
[1] Rhein Westfal TH Aachen, Dept Comp Sci, Human Language Technol & Pattern Recognit, Aachen, Germany
[2] Google Inc, New York, NY USA
来源
11TH ANNUAL CONFERENCE OF THE INTERNATIONAL SPEECH COMMUNICATION ASSOCIATION 2010 (INTERSPEECH 2010), VOLS 1-2 | 2010年
关键词
WFST; LVCSR; MINIMIZATION;
D O I
暂无
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
This paper describes a new method for building compact context-dependency transducers for finite-state transducer-based ASR decoders. Instead of the conventional phonetic decision-tree growing followed by FST compilation, this approach incorporates the phonetic context splitting directly into the transducer construction. The objective function of the split optimization is augmented with a regularization term that measures the number of transducer states introduced by a split. We give results on a large spoken-query task for various n-phone orders and other phonetic features that show this method can greatly reduce the size of the resulting context-dependency transducer with no significant impact on recognition accuracy. This permits using context sizes and features that might otherwise be unmanageable.
引用
收藏
页码:218 / +
页数:2
相关论文
共 11 条