基于3维空间Viterbi算法的汉语连续语音识别方法

赵 力;邹采荣;吴镇扬

电子学报 ›› 2000, Vol. 28 ›› Issue (7) : 84-87.

PDF(167 KB)
PDF(167 KB)
电子学报 ›› 2000, Vol. 28 ›› Issue (7) : 84-87.
论文

基于3维空间Viterbi算法的汉语连续语音识别方法

  • 赵 力, 邹采荣, 吴镇扬
作者信息 +

Recognition of Chinese Continuous Speech Based on 3-Dimension Viterbi Search

  • ZHAO Li, ZOU Cai-rong, WU Zhen-yang
Author information +
文章历史 +

摘要

本文提出了基于3维空间Viterbi算法的汉语连续语音识别方法.本方法采用60个音素单位的隐马尔可夫模型(HMM)和8个声调单位的HMM作为识别用基元模型.音素基元模型和声调基元模型的识别结果的统合,采用音素单位的HMM状态、声调单位的HMM状态和时间的3维空间Viterbi算法来实现.语音声学处理和语音言语处理的结合,采用修改型Earley分析法的Top-Done型文法分析器和One Pass DP为基础的帧同步识别算法来实现.在由10名话者发音的有关旅馆预约指南的识别困难度是27.3的1070句子的识别实验中,总平均识别率达到94.4%.

Abstract

This paper presents a recognition method of Chinese continuous speech,in which phonetic and prosodic features are integrated in terms of 3-Dimension Viterbi search.The phonetic information is modeled by 60 phonemic HMMs and the prosodic information by 8 tone HMMs.Both recognitions are synchronized based on 3-Dimension Viterbi search.A frame-synchronous parsing algorithm for CFG based on a top-down strategy is used for parsing processing.The task is related to the hotel reservation process,of which the perplexity is 27.3.For 1070 utterances produced by each of ten speakers,the average sentence recognition rate was 94.4%.

关键词

汉语连续语音识别 / 音素单位HMM / 声调单位HMM / 3维空间Viterbi算法

Key words

recognition of Chinese continuous speech / phonemic HMM / tone HMM / 3-Dimension Viterbi search

引用本文

导出引用
赵 力;邹采荣;吴镇扬. 基于3维空间Viterbi算法的汉语连续语音识别方法[J]. 电子学报, 2000, 28(7): 84-87.
ZHAO Li;ZOU Cai-rong;WU Zhen-yang. Recognition of Chinese Continuous Speech Based on 3-Dimension Viterbi Search[J]. Acta Electronica Sinica, 2000, 28(7): 84-87.
中图分类号: TP391.42    TN912.3   

基金

国家自然科学基金资助课题
PDF(167 KB)

2072

Accesses

0

Citation

Detail

国家自然科学基金资助课题
段落导航
相关文章

/