[an error occurred while processing this directive]
���¿��ټ��� �߼�����
   ��ҳ  �ڿ�����  ��ί��  Ͷ��ָ��  �ڿ�����  ��������  �� �� ��  ��ϵ����
�������պ����ѧѧ�� 2001, Vol. 27 Issue (2) :146-149    DOI:
���� ����Ŀ¼ | ����Ŀ¼ | ������� | �߼����� << | >>
��ǿ1, ëʿ��1, ����Ϊ2*
1. �������պ����ѧ ���ӹ���ϵ;
2. ���ش�ѧ ��Ϣ��ѧ�о���
Smoothed-unit HMM Algorithm in Mandarin Speech Recognition
HE Qiang1, MAO Shi-yi1, ZHANG You-wei2*
1. Beijing University of Aeronautics and Astronautics, Dept. of Electronic Engineering;
2. Wuyi University, Research Institute of Information Science

Download: PDF (0KB)   HTML 1KB   Export: BibTeX or EndNote (RIS)      Supporting Info
ժҪ ��������ʶ��Ļ�����Ԫһ��Ϊ���ء������Լ�����ĸ.������ĸΪ��Ԫ��ʶ��ϵͳ����HMMģ�ͽ���,������С,�ʺ���ʵʱʵ��.��������ģ�ͱȽϹ���,�������źŵ���ѧ���������ò�����ȷ,���ʶ����һ������ڻ�Ԫ��ϵͳ��.�������ڡ�����(tri-phone��di-phone)Ϊ��Ԫ��ϵͳ����HMMģ�������ࡢѵ����ʶ������м��������ȱ��,Ӱ�쵽ϵͳ��ʵʱ��.���������һ�����Եķ���,ϵͳ��Ԫ��ѡ������ĸ,����HMMѵ���׶�,�������������еIJ�����������,ʹ���Ϲ��ɶε�״̬�õ�ƽ��,ͬʱ���㲢����ÿ����������֮���ת�Ƹ���,ʶ��ʱ��̬��װΪ����������HMM.�÷����ڱ�������HMM������ͬʱ,�ܹ�����������,�ʺ�����DSPΪ���ĵ�ʵʱ���Ӵ�����ʶ��ϵͳ.
Email Alert
�ؼ����� ����ʶ��   ʵʱ   ����·����   HMM   ���ϻ�Ԫ   ƽ��     
Abstract�� The base unit in mandarin speech recognition is phoneme, semi-syllable or syllable. Semi-syllable system has fewer HMM models and needs less computation, thus it's suitable for real-time systems. But due to poor description for the acoustic properties of the speech signal, it generally shows a low performance compared with syllable system. While the system based on syllable or phoneme (tri-phone or di-phone) has much more HMM models, and needs massive computation in training and recognition, which goes against to real-time implementation. The new scheme is a compromised one. The new system is based on semi-syllable system, but the parameters of the entire syllable are used in training phase, so smoothing between two semi-syllable units is introduced. The transition probability between semi-syllables is calculated, and the two semi-syllable HMMs are connected into a full syllable HMM in recognition phase. This can increase the system performance without increasing HMM models, and it's fit for real-time systems with DSP kernel.
Keywords�� speech recognition   real time   Markov processes   HMM   semi-syllable unit   smoothing     
Received 1999-11-12;


About author: �� ǿ(1972-),��,�����,��ʿ��,100083,����.
��ǿ, ëʿ��, ����Ϊ.��������ʶ���ƽ�����ϻ�ԪHMM�㷨[J]  �������պ����ѧѧ��, 2001,V27(2): 146-149
HE Qiang, MAO Shi-yi, ZHANG You-wei.Smoothed-unit HMM Algorithm in Mandarin Speech Recognition[J]  JOURNAL OF BEIJING UNIVERSITY OF AERONAUTICS AND A, 2001,V27(2): 146-149
http://bhxb.buaa.edu.cn//CN/     ��     http://bhxb.buaa.edu.cn//CN/Y2001/V27/I2/146
Copyright 2010 by �������պ����ѧѧ��