基于音节时间长度高斯拟合的汉语音节切分方法

doi:10.11772/j.issn.1001-9081.2016.05.1410

计算机应用 ›› 2016, Vol. 36 ›› Issue (5): 1410-1414.DOI: 10.11772/j.issn.1001-9081.2016.05.1410

基于音节时间长度高斯拟合的汉语音节切分方法

张扬, 赵晓群, 王缔罡

同济大学电子与信息工程学院, 上海 201804

收稿日期:2015-11-13 修回日期:2016-01-10 出版日期:2016-05-10 发布日期:2016-05-09
通讯作者: 赵晓群
作者简介:张扬(1989-),男,山东滨州人,博士研究生,主要研究方向:语音信号处理;赵晓群(1962-),男,黑龙江依安人,教授,博士,主要研究方向:语音信号处理、编码理论;王缔罡(1988-),男,上海人,博士研究生,主要研究方向:编码理论。

Chinese speech segmentation method based on Gauss distribution of time spans of syllables

ZHANG Yang, ZHAO Xiaoqun, WANG Digang

College of Electronics and Information Engineering, Tongji University, Shanghai 201804, China

Received:2015-11-13 Revised:2016-01-10 Online:2016-05-10 Published:2016-05-09

摘要/Abstract

摘要： 研究汉语自然语音音节切分方法具有明显现实意义,比较准确的自然语音切分方法可以代替人工对一些拥有参照文本的语音进行标注。然而至今为止并没有完全准确的汉语语音音节切分方法。依据相同发音环境下汉语语音音节时间长度服从某种高斯分布和相邻语音音节之间存在短时能量波谷两个假设,提出了基于音节时间长度高斯拟合的汉语音节切分方法。对算法进行分析,根据初步切分短时能量波谷分散到各分语音段的特性,提出了简化算法,有效降低了该音节切分方法的时间复杂度。实验结果表明,音节切分准确度(与人工标注切分时间距离平方的均值)达到小数点后3位,在台式机Matlab环境下运算时间均不超过1 s,可以达到应用要求。

关键词: 汉语, 自然语音, 音节切分, 时间长度, 波谷, 高斯分布

Abstract: So far away, there is no accurate method for Chinese natural speech segmentation of syllables,which is meaningful in labeling speech with reference text instead of people. According to two hypotheses that time spans of Chinese syllables under the same pronunciation obey Gauss distribution and short-time energy valley exists between two adjacent syllables, Chinese speech segmentation method based on Gauss distribution of time spans of syllables was proposed. A simplified method based on distribution of energy valleys was given, which effectively reduced the time complexity of this speech segmentation method. The experimental results show that segmentation accuracy (mean square value of time spans between artificial labels and labels created by this method) achieve 10^-3 and computing times are less than 1 s in Matlab of PC.

Key words: Chinese, natural speech, speech segmentation, time span, valley, Gauss distribution

中图分类号:

TP391.4

张扬, 赵晓群, 王缔罡. 基于音节时间长度高斯拟合的汉语音节切分方法[J]. 计算机应用, 2016, 36(5): 1410-1414.

ZHANG Yang, ZHAO Xiaoqun, WANG Digang. Chinese speech segmentation method based on Gauss distribution of time spans of syllables[J]. Journal of Computer Applications, 2016, 36(5): 1410-1414.

参考文献

[1] TOLEDANO D T, GOMEZ L A H, GRANDE L V. Automatic phonetic segmentation[J]. IEEE Transactions on Speech and Audio Processing, 2003, 11(6):617-625.
[2] WU Y J, KAWAI H, NI J, et al. Discriminative training and explicit duration modeling for HMM-based automatic segmentation[J]. Speech Communication, 2005, 47(3):397-410.
[3] van HEMERT J P. Automatic segmentation of speech[J]. IEEE Transactions on Signal Processing, 1991, 39(4):1008-1012.
[4] CHOU F C, TSENG C Y, LEE L S. A set of corpus-based text-to-speech synthesis technologies for Mandarin Chinese[J]. IEEE Transactions on Speech and Audio Processing, 2002, 10(7):481-494.
[5] 杜守栓. 方言口音普通话语音自动切分算法研究[D].北京:中国科学院, 2006:15-26.(DU S S. Research on robust automatic segmentation of dialectal speech[D]. Beijing:University of Chinese Academy of Sciences, 2006:15-26.)
[6] 何可嘉. 广播语言的自动标注系统[D].北京:北京邮电大学, 2010:22-47.(HE K J. An automatic labeling system for broadcast news[D]. Beijing:Beijing University of Posts and Telecommunications, 2010:22-47.)
[7] 韩虎. 汉语连续语音的音节自动标注算法研究及实现[D].哈尔滨:哈尔滨工业大学, 2008:21-44.(HAN H. Research and realization of the automatic syllable marking algorithm for Chinese continuous speech[D]. Harbin:Harbin Institute of Technology, 2008:21-44.)
[8] LEE K S. MLP-based phone boundary refining for a TTS database[J]. IEEE Transactions on Audio, Speech and Language Processing, 2006, 14(3):981-989.
[9] BROGNAUX S, DRUGMAN T. HMM-based speech segmentation:improvements of fully automatic approaches[J]. IEEE Transactions on Audio, Speech and Language Processing, 2016, 24(1):5-15
[10] 廖文辉, 刘炎.数据分析与SAS实验[M].北京:经济科学出版社, 2010:13-32.(LIAO W H, LIU Y. Data Analysis and SAS Experiment[M]. Beijing:Economic Science Press, 2010:13-32.)
[11] 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, 2013:117-129.(SONG Z Y. Application of Matlab in Speech Signal Analysis and Synthesis[M].Beijing:Beihang University Press, 2013:117-129.)
[12] 章森, 刘磊, 刁麓弘.大规模语音语料库及其在TTS中应用的几个问题[J].计算机学报, 2010, 33(4):667-696.(ZHANG S, LIU L, DIAO L H. Problems on large-scale speech corpus and the applications in TTS[J]. Chinese Journal of Computers, 2010, 33(4):667-696.)BackgroundZHANG Yang, born in 1989, Ph. D. candidate. His research interests include speech signal processing.ZHAO Xiaoqun, born in 1962, Ph. D., professor. His research interests include speech signal processing, coding theory.WANG Digang, born in 1988, Ph. D. candidate. His research interests include coding theory.

[1]	李灿, 杨雅婷, 马玉鹏, 董瑞. 基于语种相似性挖掘的神经机器翻译语料库扩充方法[J]. 《计算机应用》唯一官方网站, 2021, 41(11): 3145-3150.
[2]	王岩, 罗倩, 邓辉. 基于Gibbs抽样的轴承故障诊断方法[J]. 计算机应用, 2018, 38(7): 2136-2140.
[3]	李龙澍, 翁晴晴. 基于反向学习的自适应差分进化算法[J]. 计算机应用, 2018, 38(2): 399-404.
[4]	张扬, 赵晓群, 王缔罡. 基于时频二维能量特征的汉语音节切分方法[J]. 计算机应用, 2016, 36(11): 3222-3228.
[5]	王荣波, 李杰, 黄孝喜, 周昌乐, 谌志群, 王小华. 基于多元判别分析的汉语句群自动划分方法[J]. 计算机应用, 2015, 35(5): 1314-1319.
[6]	王希杰. 词位标注汉语分词中上下文有效范围定量分析[J]. 计算机应用, 2012, 32(05): 1340-1342.
[7]	曲怀敬. 基于互补特征的纹理图像检索[J]. 计算机应用, 2012, 32(04): 1101-1103.
[8]	韩玉峰王小林. 基于主动形状模型算法的局部灰度模型的加权改进方法[J]. 计算机应用, 2011, 31(12): 3392-3394.
[9]	黄儿松刘金华文汝红. 基于小波视觉模型的乘性水印算法[J]. 计算机应用, 2011, 31(08): 2165-2168.
[10]	路漫漫滕奇志. 蚁群算法实现的图像边缘连接[J]. 计算机应用, 2010, 30(4): 932-934.
[11]	武海洋王慧程宝琴. 基于最小Bayes风险的小波域局部自适应图像去噪[J]. 计算机应用, 2010, 30(12): 3238-3240.
[12]	王改良武妍. 基于仿生模式识别理论的声调识别[J]. 计算机应用, 2010, 30(10): 2709-2711.
[13]	吕浩音. 可训练文语转换系统的时长模型优化[J]. 计算机应用, 2010, 30(1): 282-284.
[14]	王益艳. 基于广义变分模型的自适应图像去噪算法[J]. 计算机应用, 2009, 29(11): 3033-3036.
[15]	张立晏琦. 基于高斯分布和模拟退火算法的免疫微粒群优化算法研究[J]. 计算机应用, 2008, 28(9): 2392-2394.

基于音节时间长度高斯拟合的汉语音节切分方法

Chinese speech segmentation method based on Gauss distribution of time spans of syllables

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics