新疆大学信息科学与工程学院
纸质出版:2015
移动端阅览
[1]宋洋,努尔买买提·尤鲁瓦斯,吾守尔·斯拉木.维吾尔语韵律调节研究[J].新疆大学学报(自然科学版),2015,32(04):453-461.
[1]宋洋,努尔买买提·尤鲁瓦斯,吾守尔·斯拉木.维吾尔语韵律调节研究[J].新疆大学学报(自然科学版),2015,32(04):453-461. DOI: 10.13568/j.cnki.651094.2015.04.011.
DOI:10.13568/j.cnki.651094.2015.04.011.
维吾尔语音韵律的调节要充分考虑其独特的语言文化
本文研究维吾尔语音的韵律特征并结合重音、停顿等语言现象对维吾尔语的韵律实现调节.本文录制维吾尔语情感语料
应用ANN(Artificial Neural Network)对停顿进行建模计算;重音的调节则依据维吾尔语的词法规则
整理出重音异义词本
将Fujisaki模型应用在局部单词的层面上
更准确的调节重音出现的位置;最后对提取基音频率的方法作改进
提出含调节参数的LPC
实现对语音基音频率曲线的调节.经过测试
模型停顿的计算预判准确率可以达到71.2%
对重音的调节准确度可以达到86.7%
对基音频率的调节可以明显体现出情感调型.
Uyghur specializes itself with unique linguistic culture
which has to be considered when modify rhythm of Uyghur speech. This paper researches on the rhythmic features of Uyghur and references the special speech phenomenon of Uyghur. This paper records emotional Uyghur speech materials
and applies Artificial Neural Networks to predict where break downs might take place. The adjustment for stresses is based on lexical of Uyghur. This paper summaries a stressing-homograph word list. Fujisaki is used to portray stressing line in syllable level LPC gets modified by introducing two parameters as LPCAP that plays a part in pitch frequency adjustment. According to the results of the test
the accuracy of prediction in break downs can be 71.2%
accuracy in stress adjustment can be 86.7%; the adjustments in pitch frequency can efficiently perform emotional tones.
胡航.现代语音信号处理[M].北京:电子工业出版社,2014:7.
麦麦提艾力·吐尔逊,吾守尔·斯拉木.维吾尔语拼接式语音合成方法研究[J].新疆大学学报,2006,33(1):202-203.
孜丽卡木·哈斯木,那斯尔江·吐尔逊,吾守尔·斯拉木.维吾尔语词首音节元音声学分析[J].中文信息学报,2009,5(23):114-118.
王辉,努尔买买提·尤鲁瓦斯,吾守尔·斯拉木.维吾尔语音素的声学特征分析[J].中文信息学报,2014,28(1):101-106.
阿依提拉·米吉提.维吾尔语音情感声学特征提取与建模研究[J].通信技术,2013,46(11):51-54.
曹剑芬.基于语法信息的汉语韵律结构预测[J].中文信息学报,2003(3):41-46.
He Ling,Huang Hua.Margaret Lech Emotional Speech Synthesis Based on Prosodic Feature Modification s[J].Engineering,2013:573-77.
Heiga Zen,Andrew Senior,Mike Schuster.Statistical Parametric Speech Synthesis Using Deep Neural Networks[A].IEEE.@google.com.
Sudhaka Sangeetha,Sekar Jothilakshmi.Syllable Based Text to Speech Synthesis System Using Auto Associative Neural Network Prosody Prediction[J].International Journal of Speech Technology,2014,17(2):91-98.
陈宝林.最优化理论与算法[M].北京:清华大学出版社,2005:10.
聂晓丽.Fujisaki模型在维吾尔语语音合成中的应用[J].电声技术,2007,31(7):51-55.
Ranniery Maia,Masami Akamine,Mark J F Gales.Complex Cepstrum for Statistical Parametric Speech Synthesis[J].Speech Communication,2013,55:606-618.
地理木拉提·吐尔逊,古丽娜尔·艾力,热娜古丽·达古提,等.维吾尔语陈述句语调的起伏度[J].清华大学学报,2011,51(9),1191-1195.
Chiu Yu Tseng,Shao Huang Pin,Yehlin Lee,et al.Fluent Speech Prosody:Framework and modeling[J].Speech Communication,2005,46:284-309.
宋知用.MATLAB在语音信号分析与合成中的应用[M].北京:北京航空航天大学出版社,2013:11.
0
浏览量
124
下载量
0
CSCD
关联资源
相关文章
相关作者
相关机构
京公网安备11010802024621
