摘要
针对智能化计算机中汉语语音识别这一重要研究课题,利用小波分析方法,以神经网络为基础,提出了基于能量的汉语语音音节划分的自动实现的方法,这是实现汉语语音识别的基础和前提。
To counter the problem of Chinese speech recognition as an important one for intellectualized computer by wavelet analysis and neural network theory a method of automatic realizing of energy-based Chinese speech syllable partition is given, which is the foundation and the precondition of realizing Chinese speech recognition.
出处
《哈尔滨理工大学学报》
CAS
1999年第6期1-4,共4页
Journal of Harbin University of Science and Technology
基金
黑龙江省自然科学基金(900269)
关键词
音节
短时能量
语音识别
汉语
自动划分
syllable
short-time interval across zero
short-time energy
speech recognition
