分级识别的策略在模式识别领域中提出相当长的时间了。尽管人类可以训练地使用这个策略进行识别,但对语音识别而言,缺少一个有效的系统化的方法来实现它。本文给出了我们最近在这方面做的一些研究工作,使用了子空间划分原理来实现一个分级识别器,并用树型结构来组织多个识别器。实验结果表明,该方法与传统方法相比,误识率降低10%。我们将在未来的研究工作中,测试全部汉语音节,并将该方法扩展到连续语音识别。
Abstract
Hierarchical recognition has been proposed for a long time in the pattern recognition field. Although it is a familiar action when human performs a recognition task , there is not an effective and systematic method to implement it for the speech recognition. This paper presents our recent experimental results on this topic , which uses the principle of sub-space partition to realize a hierarchical recogntion and a tree-based architecture to organize multi-recognizers. The results show that the proposed algorithm can achieve about 10% error reduction compared with traditional methods. In future works , we will test all Chinese syllables and extend them for the continous speech recogntion.
关键词
计算机应用 /
中文信息处理 /
语音识别 /
分级识别 /
空间划分
{{custom_keyword}} /
Key words
computer application /
Chinese information processing /
speech recognition /
hierarchical recognition /
space partition
{{custom_keyword}} /
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] Breiman , L. , et al. , Classification and Regression Trees [M] , Pacific Grove , CA , Wadsworth , 1984.
[2] Quinlan , J. R. , Introduction of Decision Trees , in Machine Learning : An Artifical Intelligence Approach [M] . Boston , Kluwer Academic Publishers , 1986 , 1 - 86.
[3] Rogova ,G. , Combining the Results of Several Neural Network Classifiers [J] , Neural Network , 1994 , 7 (5) : 777 - 781.
[4] Ho , T. , Hull , J. and Srihari , S. , Decision Combination in Multiple Classifiers Systems [J] , IEEE Trans. on Pattern Analysis and Machine Intelligence , Jan. 1994 , 16 (1) : 66 - 75.
[5] Lam , L. and Suen , C.-Y. , Optimal Combinations of Pattern Classifiers [J] , Pattern Recognition Letters , 1995 , 16 : 945 - 954.
[6] Kevin Woods , W. Philip Kegelmeyer and Kevin Bowyer , Combination of Multiple Classifiers Using Local Accuracy Estimates [J] , IEEE Trans. on Pattern Analysis and Machine Intelligence , Apr. 1997 , 19 (4) : 405 - 409.
[7] Halberstadt , A. and Glass , J. Hetergeneous Measurements for Phonetic Classification [A] , Proc. EUROSPEECH [C] , 401 - 404 , 1997.
[8] Halberstadt , A. and Glass , J. Heterogeneous Measurements and Multiple Classifiers for Speech Recognition [A] , Proc. ICSLP [C] , Sydney , Australia : 1998.
[9] Yang , D.-L. , Xu , M.-X. , Wu , W.-H. Study on the Strategy for Hierarchical Speech Recognition [A] , International Symposium on Chinese Spoken Language Processing [C] , Taipei : 2002.
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}
基金
国家留学基金委资助的中德重点实验室合作项目“基于复杂上下文感知的数字助理”支持
{{custom_fund}}
PDF(198 KB)
