论文部分内容阅读
利用时频分析技术中的重排方法对中文普通话发音中得六种元音声调进行分析,提出了符合语音学中的声调格局的时频特征,实验结果表明,利用这一特征,对提高中文语音识别的效率和中文语音识别的成熟有直接的促进作用。语音格局一般分为元音格局、辅音格局、语调格局等。汉语等声调语言还有声调格局[1]。人们对汉语发音的研究往往以其声调的分析作为开端。而语音作为一种携带信息的信号,是通过调节其信号频率
Using the rearrangement method of time-frequency analysis technique, we analyze the six kinds of vowel tones in Chinese Mandarin pronunciation and present the time-frequency characteristics in accordance with the tone pattern in phonetics. The experimental results show that using this feature, The efficiency of speech recognition and Chinese speech recognition has a direct role in promoting. Voice patterns are generally divided into vowel pattern, consonant pattern, intonation pattern. Tone tone languages such as Chinese tone pattern [1]. People often begin their research on the pronunciation of Chinese in terms of their tone. The voice as a signal carrying information through the regulation of its signal frequency