Computer Engineering and Applications ›› 2008, Vol. 44 ›› Issue (2): 224-226.

• 工程与应用 • Previous Articles     Next Articles

New keyword spotting model based on layered query sheet

AN Zhen-zhou1,2,YANG Jian1,QIU Wen2   

  1. 1.School of Information Science and Technology,Yunnan University,Kunming 650091,China
    2.Department of Computer Science,Yuxi Normal University,Yuxi,Yunnan 653100,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2008-01-11 Published:2008-01-11
  • Contact: AN Zhen-zhou

一种新的基于分层查询表的关键词识别模型

安镇宙1,2,杨 鉴1,仇 汶2   

  1. 1.云南大学 信息学院,昆明 650091
    2.玉溪师范学院 计算机科学系,云南 玉溪 653100
  • 通讯作者: 安镇宙

Abstract: Keyword spotting is a hotspot of speech recognition research in recent years.This paper proposes a new keyword spotting model based on layered query sheet and this model is simple,applied and fast.The author uses this model to realize the road information query system and this system has high recognition accuracy and practicability.

Key words: keyword spotting, layered query, feature space trace, parallel DTW

摘要: 关键词识别是近年来语音识别研究的一个热点。提出了一种新的基于分层查询表的关键词识别模型,该模型具有简单、实用、快速的特点。利用该模型实现了路况信息查询系统,取得了较高的识别率,具有一定的实用性。

关键词: 关键词识别, 分层查询, 特征空间轨迹, 并行DTW