藏语拉萨话大学生的多维嗓音分析
发布时间：2019-06-14 01:48
【摘要】：藏语拉萨话是藏语三大方言卫藏方言、康方言、安多方言中的卫藏方言,本文按照年龄分段,将18-25岁的以藏语拉萨话为母语的大学生为研究对象,通过MDVP即多维嗓音分析软件,按照男女生的性别,分别进行嗓音的特性研究,并进行比较分析,来揭示藏语拉萨话的多维嗓音的特性。MDVP,即多维嗓音分析软件,由美国Kay公司生产,最早是从嗓音病理领域发展出来的一种通过声音检测嗓音质量、发声类型和诊断嗓音病变的声学方法。首先,对藏语拉萨话的声母进行多维嗓音的分析,按声带是否震动来分的话,大部分的声母是清辅音,而清辅音嗓音不振动,无法得到具有功能意义的参数进行分析,浊声母声带颤动相对清声母来说就非常明确了,并能够获得足够的嗓音参数进行计算分析。藏语拉萨话浊声母有:m、n、(?)、(?)、l、w、r、j八个,本文主要分析藏语拉萨话男女声的浊声母的多维嗓音分析。藏语拉萨话男女生浊声母的嗓音参数基频FO在平均值上女声远高于男声,最高基频Fhi和最低基频Flo的平均值数据都是女声高于男声;分析样本时长、音调扰动商的参数数据,不管是平均值还是标准差,男女生两者的参数数据都是没有很大差值,基本相同。其次,利用MDVP多维嗓音分析软件提取6类33项嗓音参数:第一类,基音基础参数;第二类,频率抖动参数;第三类,振幅抖动参数;第四类,嗓音指数;第五类,嗓音清化参数;第六类,基本参数对藏语拉萨话的韵母进行多维嗓音的分析。基音基础类参数中最高基频女声要高于男声,平均音调周期和基频男女声呈一个倒数关系;藏语拉萨话的男女声韵母都是随着基频的提高其基频半音范围也在随之提高。嗓音指数类中藏语拉萨话男女生韵母的清浊率和嗓音骚动的数据变化性质相同,且两者的数据也没有大的差距。嗓音清华参数类中藏语拉萨话男女生韵母的清声级DUV的数值比藏语拉萨话男女生韵母的非浊音段数高。最后,藏语拉萨话声调男女声在一方面提取音高、开商和速度商三类,把所有采到的数据中提取所需的参数,然后再进行平均计算。藏语拉萨话声调的男女生嗓音的平均音高类中,音高范围和最大音高、音高标准偏差、音高周期这4类女声要比男声高;而平均音高和最小音高、音高抖动则相反,男声要比女声高;藏语拉萨话声调的男女生嗓音开商的数值差异较大,且开商的平均速度随着音高的提高而下降,呈现出反比关系;藏语拉萨话声调的男女生嗓音速度商的数值差异较大,表示藏语拉萨话声调的男女生嗓音在速度商方面有很大的差别。
[Abstract]:Tibetan Lhasa dialect is the Wei Zang dialect in Wei Zang dialect, Kang dialect and Ando dialect. According to the age segment, this paper studies the voice characteristics of Tibetan Lhasa dialect by MDVP, that is, multi-dimensional voice analysis software, according to the sex of male and female students, and makes a comparative analysis to reveal the multi-dimensional voice characteristics of Tibetan Lhasa dialect. Multi-dimensional voice analysis software, produced by Kay Company of the United States, was first developed from the field of voice pathology to detect voice quality, vocal type and diagnosis of voice lesions. First of all, the multi-dimensional voice analysis of Tibetan Lhasa consonants is carried out. According to whether the vocal cord vibrates or not, most of the consonants are clear consonants, but the clear consonant voice does not vibrate, so it is impossible to get the functional parameters to analyze the consonants. The vocal cord vibration is very clear compared with the clear consonants, and enough voice parameters can be obtained for calculation and analysis. There are eight vocal initials in Tibetan Lhasa dialect: M, n, (?), l, w, r, j. This paper mainly analyzes the multi-dimensional voice analysis of the turbid consonants of Tibetan Lhasa men and women. The average value of female voice parameter FO is much higher than that of male voice in Tibetan Lhasa, and the average data of highest fundamental frequency Fhi and lowest fundamental frequency Flo are higher than that of male voice. The time of analysis sample, the parameter data of tone disturbance quotient, whether average or standard deviation, there is no great difference between male and female, and the parameter data of male and female students are basically the same. Secondly, using MDVP multi-dimensional voice analysis software to extract 6 kinds of 33 voice parameters: the first type, pitch basic parameters; the second type, frequency jitter parameters; the third type, amplitude jitter parameters; the fourth type, voice index; the fifth type, voice clearing parameters; the sixth type, the basic parameters to analyze the Tibetan Lhasa vowels. Among the basic parameters of pitch, the highest fundamental frequency female voice is higher than that of male voice, and the average tone period is a countdown to the fundamental frequency male and female sound, and the male and female phonemes in Tibetan Lhasa dialect are also increasing with the increase of fundamental frequency. In the voice index category, the turbid rate of male and female rhymes in Tibetan Lhasa dialect is the same as the data of voice disturbance, and there is no big difference between the two data. In the voice Tsinghua parameter class, the value of the clear sound level DUV of the male and female vowels in Tibetan Lhasa dialect is higher than that of Tibetan Lhasa male and female vowels. Finally, the male and female tones of Tibetan Lhasa dialect are extracted from pitch, quotient and speed quotient on the one hand, and the required parameters are extracted from all the collected data, and then the average calculation is carried out. In the average pitch category of boys and girls in Tibetan Lhasa tone, the pitch range and maximum pitch, the standard deviation of pitch, and the pitch cycle of these four kinds of female voices are higher than male voices; while the average pitch and minimum pitch, pitch jitter are opposite, the male voice is higher than the soprano, and the average speed of business opening decreases with the increase of the pitch, showing an inverse relationship. There is a great difference in the voice speed quotient of boys and girls in Tibetan Lhasa tone, which indicates that there is a great difference in speed quotient between boys and girls in Tibetan Lhasa tone.
【学位授予单位】：西北民族大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：H214
本文编号：2498998
[Abstract]:Tibetan Lhasa dialect is the Wei Zang dialect in Wei Zang dialect, Kang dialect and Ando dialect. According to the age segment, this paper studies the voice characteristics of Tibetan Lhasa dialect by MDVP, that is, multi-dimensional voice analysis software, according to the sex of male and female students, and makes a comparative analysis to reveal the multi-dimensional voice characteristics of Tibetan Lhasa dialect. Multi-dimensional voice analysis software, produced by Kay Company of the United States, was first developed from the field of voice pathology to detect voice quality, vocal type and diagnosis of voice lesions. First of all, the multi-dimensional voice analysis of Tibetan Lhasa consonants is carried out. According to whether the vocal cord vibrates or not, most of the consonants are clear consonants, but the clear consonant voice does not vibrate, so it is impossible to get the functional parameters to analyze the consonants. The vocal cord vibration is very clear compared with the clear consonants, and enough voice parameters can be obtained for calculation and analysis. There are eight vocal initials in Tibetan Lhasa dialect: M, n, (?), l, w, r, j. This paper mainly analyzes the multi-dimensional voice analysis of the turbid consonants of Tibetan Lhasa men and women. The average value of female voice parameter FO is much higher than that of male voice in Tibetan Lhasa, and the average data of highest fundamental frequency Fhi and lowest fundamental frequency Flo are higher than that of male voice. The time of analysis sample, the parameter data of tone disturbance quotient, whether average or standard deviation, there is no great difference between male and female, and the parameter data of male and female students are basically the same. Secondly, using MDVP multi-dimensional voice analysis software to extract 6 kinds of 33 voice parameters: the first type, pitch basic parameters; the second type, frequency jitter parameters; the third type, amplitude jitter parameters; the fourth type, voice index; the fifth type, voice clearing parameters; the sixth type, the basic parameters to analyze the Tibetan Lhasa vowels. Among the basic parameters of pitch, the highest fundamental frequency female voice is higher than that of male voice, and the average tone period is a countdown to the fundamental frequency male and female sound, and the male and female phonemes in Tibetan Lhasa dialect are also increasing with the increase of fundamental frequency. In the voice index category, the turbid rate of male and female rhymes in Tibetan Lhasa dialect is the same as the data of voice disturbance, and there is no big difference between the two data. In the voice Tsinghua parameter class, the value of the clear sound level DUV of the male and female vowels in Tibetan Lhasa dialect is higher than that of Tibetan Lhasa male and female vowels. Finally, the male and female tones of Tibetan Lhasa dialect are extracted from pitch, quotient and speed quotient on the one hand, and the required parameters are extracted from all the collected data, and then the average calculation is carried out. In the average pitch category of boys and girls in Tibetan Lhasa tone, the pitch range and maximum pitch, the standard deviation of pitch, and the pitch cycle of these four kinds of female voices are higher than male voices; while the average pitch and minimum pitch, pitch jitter are opposite, the male voice is higher than the soprano, and the average speed of business opening decreases with the increase of the pitch, showing an inverse relationship. There is a great difference in the voice speed quotient of boys and girls in Tibetan Lhasa tone, which indicates that there is a great difference in speed quotient between boys and girls in Tibetan Lhasa tone.
【学位授予单位】：西北民族大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：H214
【相似文献】
相关会议论文 前2条
1 赵知劲;袁慧琴;郭殿武;;基于Hilbert-Huang的信号瞬时参数提取[A];第十二届全国信号处理学术年会（CCSP-2005）论文集[C];2005年
2 汪洁;孙玲玲;刘军;刘智力;;微波HBT小信号等效电路的参数提取[A];2003'全国微波毫米波会议论文集[C];2003年
相关重要报纸文章 前1条
1 赵文恒;A口参数提取保证网络优化实施[N];通信产业报;2002年
相关博士学位论文 前2条
1 孙华;杉木人工林林分主要参数遥感反演技术研究[D];中国林业科学研究院;2013年
2 沈溧;高电子迁移率晶体管微波建模与参数提取研究[D];华东师范大学;2016年
相关硕士学位论文 前10条
1 刘金鹏;基于地面激光扫描数据的树木参数提取研究[D];中国林业科学研究院;2016年
2 赵军;人体组织偏振参数高分辨成像方法与系统研究[D];南京理工大学;2017年
3 才项措;藏语拉萨话大学生的多维嗓音分析[D];西北民族大学;2017年
4 冯松;深亚微米全耗尽SOI MOSFET参数提取方法的研究[D];西安理工大学;2007年
5 郑文思;藏语夏河话文本驱动唇形参数合成系统[D];西北民族大学;2013年
6 朱辉;亚90纳米CMOS器件PSP建模技术和参数提取的研究[D];上海交通大学;2008年
7 丁猛;GaN HEMT小信号等效电路参数提取[D];西安电子科技大学;2009年
8 方敏;矩阵束法的改进及其应用[D];苏州大学;2007年
9 李广伟;基于船载LiDAR的湖沿岸植被参数估测[D];首都师范大学;2012年
10 辛慧杰;视频图像中阴影的去除和交通参数提取的研究及实现[D];东北师范大学;2007年
，本文编号：2498998
本文链接：https://www.wllwen.com/wenyilunwen/yuyanyishu/2498998.html
