摘要
根据语音合成与识别等语音应用研究的需求,从文本分析模块入手,从"维吾尔语语音声学参数库"中选择了包含边音/l/的单音节、双音节、三音节以及多音节词[0](即四音节以及四音节以上的词),提取它的声学参数并进行统计分析,探讨了边音的声学特性,分别研究了边音在单音节及多音节词中时的以及边音出现在词中位置(词首、词中、词末)时的共振峰、音强和时长分布模式,得出了一系列结论,其目的是为了提高语音合成的自然度即更好地为自然语言处理服务。本项研究结果维吾尔语语言乃至整个阿尔泰语系语言的韵律研究具有较高的参考价值。
Considering the requirements of speech applications such as speech synthesis and speech recognition,in order to improve the naturalness of speech synthesis system,this paper have done the systematic empirical work of acoustic feature analysis on the different structured(single syllable,two-syllable,three syllable and more than three syllable)words which contains lateral /l/ in Uyghur language,and get some valuable conclusions.The experimental analysis was carried out on the basis of Uyghur Acoustic Database.We believe that the study of this project not only has high research value for further studying of the Uyghur natural language,but also has high research value for the study of entire Altay language family.
出处
《信息通信》
2011年第2期31-34,共4页
Information & Communications
基金
国家自然科学基金(61063023)
新疆维吾尔自治区自然科学基金资助(200821125)
关键词
语音
维吾尔语
边音
时长
音强
共振峰
Speech
Uyghur language
Lateral
Length
Intensity
Formant