Parameter Tuning for Fast Speech Recognition

被引：0

作者：

Colthurst, Thomas ^{[1
]}

Arvizo, Tresi ^{[1
]}

Kao, Chia-Lin ^{[1
]}

Kimball, Owen ^{[1
]}

Lowe, Stephen ^{[1
]}

Miller, David R. H. ^{[1
]}

Van Sciver, Jim ^{[1
]}

机构：

[1] BBN Technol Inc, Cambridge, MA 02138 USA

来源：

INTERSPEECH 2007: 8TH ANNUAL CONFERENCE OF THE INTERNATIONAL SPEECH COMMUNICATION ASSOCIATION, VOLS 1-4 | 2007年

关键词：

speech recognition;

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：

We describe a novel method for tuning the decoding parameters of a speech-to-text system so as to minimize word error rate (WER) subject to an over-all time constraint. When applied to three sub-realtime systems for recognizing English conversational telephone speech, the method gave speed improvements of up to 21.1% while at the same time reducing WER by up to 6.7%.

引用

页码：2576 / 2579

页数：4