The Implementation of a Practical Chinese Speech Recognition System For The Parcel Post Checking Task

SHAN Yi-xiang; ZHANG Hao-tian; LI Hu-sheng; ZHONG Lin; ZHANG Jin; LIU Jia; LIU Run-sheng

您当前的位置：

首页 >

文章列表页 >

The Implementation of a Practical Chinese Speech Recognition System For The Parcel Post Checking Task

更新时间：2025-07-16

- The Implementation of a Practical Chinese Speech Recognition System For The Parcel Post Checking Task
- Acta Electronica Sinica Vol. 30, Issue 4, Pages: 544-547(2002)
- 作者机构：
  
  清华大学电子工程系,北京,100084
- 作者简介：
- 基金信息：
- DOI：
  CLC： TN912.3
- Published：2002
- 稿件说明：
移动端阅览
SHAN Yi-xiang, ZHANG Hao-tian, LI Hu-sheng, et al. The Implementation of a Practical Chinese Speech Recognition System For The Parcel Post Checking Task[J]. Acta Electronica Sinica, 2002, 30(4): 544-547.
DOI：

SHAN Yi-xiang, ZHANG Hao-tian, LI Hu-sheng, et al. The Implementation of a Practical Chinese Speech Recognition System For The Parcel Post Checking Task[J]. Acta Electronica Sinica, 2002, 30(4): 544-547. DOI：

摘要

本文研究开发了一套邮包信息校核语音识别系统.该系统利用中大词汇量非特定人连续语音识别技术实时实现了邮包信息的语音校核.系统可以识别普通话或四川话语音

可识别的词汇量约为4500条.系统还采用了拒识技术与说话人自适应技术

提高了整个系统的稳健性.实验表明对普通话的首选识别率达到98.7%

前三选识别率达到99.9%.对四川话的首选识别率达到95.9%

前三选识别率达到98.6%

对无关语音的正确拒识率达到85%

对口音较重的说话人经过自适应后识别率可提高5－8个百分点.

Abstract

In this paper a real-time Parcel Post Checking Speech Recognition System is developed.This system can achieve the parcel post information checking and verification by means of a medium vocabulary speaker-independent continuous speech technology.Input speech can be Mandarin or Sichuan dialect.Vocabulary size is about 4500.By employing the advantages of the rejection model and speaker adaptation technology

the system's robustness is improved remarkably.The experiments show that the system has a very good performance.For mandarin voice input

the recognition accuracy of the top one candidate is 98.7%

and the recognition accuracy of the top three candidates is 99.9%; For Sichuan dialect one the recognition accuracy of the top one candidate is 95.9%

and the recognition accuracy of the top three candidates is 98.6%.

关键词

Keywords

references

Views

954

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Technology of Speaker Adaptation in Speech Recognition and Its Development Trend

High Performance Mandarin Digit String Speech Recognition

Related Author

LI Hu-sheng

LIU Jia

LIU Run-sheng

Related Institution

Department of Electronic Engineering, Tsinghua University

⁰