Prosodic Structure Prediction based on Conditional Random Field Model

doi:10.13190/jbupt.200905.36.dongy

JOURNAL OF BEIJING UNIVERSITY OF POSTS AND TELECOM ›› 2009, Vol. 32 ›› Issue (5): 36-40.doi: 10.13190/jbupt.200905.36.dongy

• Papers • Previous Articles Next Articles

Prosodic Structure Prediction based on Conditional Random Field Model

DONG Yuan1,ZHOU Tao1,DONG Cheng-yu2,WANG Hai-la2

DONG Yuan1,〓ZHOU Tao1,DONG Cheng-yu2,WANG Hai-la2

Received:2009-03-11 Revised:2009-08-03 Online:2009-10-28 Published:2009-10-28

Abstract

Abstract:

Prosodic structure prediction is an important component in mandarin text-to-speech (TTS) system. A prosodic structure prediction method is proposed, based on the conditional random field (CRF) algorithm. Prosodic word model and prosodic phrase model utilize CRF method for machine learning based on automatically segmented and tagged features and hierarchal prosodic structure information extracted from a large-scale manually labeled speech corpus. The approach achieves F-score of 90.67% in prosody word prediction and 80.05% in prosody phrase prediction, 3.62% and 5.65% higher than that of max entropy (ME) algorithm based method. Experiment results show that the approach of CRF based method makes considerable improvement in prosodic structure prediction, and works well in real mandarin TTS system.

Key words: text-to-speech, prosodic structure, conditional random field, machine learning

DONG Yuan1,ZHOU Tao1,DONG Cheng-yu2,WANG Hai-la2. Prosodic Structure Prediction based on Conditional Random Field Model[J]. JOURNAL OF BEIJING UNIVERSITY OF POSTS AND TELECOM, 2009, 32(5): 36-40.

References

[1] Chou F C, Tseng C Y, Lee L S, et al. Automatic generation of prosodic structure for high quality Mandarin speech synthesis//Proceeding of the Fourth International Conference on Spoken Language Processing. Philadelphia: , 1996: 1624-1627.
[2] Chu Min, Lu S N. A text-to-speech system with high intelligibility and high naturalness for Chinese[J]. Chinese Journal of Acoustics, 1996, 15(1): 81-90.
[3] Niu Zhengyu, Chai Peiqi. Segmentation of prosodic phrases for improving the naturalness of synthesized mandarin Chinese speech//ICSLP 2000. Beijing: , 2000: 350-353.
[4] Mao Xinnian, Dong Yuan, Han Jinyu, et al. Inequality maximum entropy classifier with character featrues for polyphone disambiguation in mandarin TTS systems//IEEE International Conference on Acoustics, Speech, and Signal Processing. Honolulu: , 2007: IV-705-IV-708.
[5] Wang M, Hirschberg J. Automatic classification of intonational phrase boundaries[J]. Computer Speech and Language, 1992, 16(6): 175-196.
[6] Taylor P, Black A W. Assigning phrase breaks from part-of-speech sequences[J]. Computer Speech and Language, 1998, 12(2): 99-117.
[7] Jia Yuxiang, Huang Dezhi, Liu Wu, et al. Text normalization in mandarin text-to-speech system//IEEE International Conference on Acoustics, Speech, and Signal Processing. Las Vegas: , 2008: 4693-4696.
[8] Lafferty J, McCallum A, Pereira F C N. Conditional random fields: probabilistic models for segmenting and labeling sequence data//Proc of the 18th ICML. San Francisco: , 2001: 282-289.
[9] Mao Xinnian, Dong Yuan, He Saike, et al. Chinese word segmentation and named entity recognition based on conditional random fields//IJCNLP 2008. Hyderabad: , 2008: 90-93.
[10] Brill E. Transformation-based error-driven learning and natural language processing: a case study in part of speech tagging[J]. Computational Linguistics, 1995, 21(4): 543-565.

[1]	CHANG Xiao, HUANG Zhibin, YU Min, YANG Wubing. A Deep Decision Tree Model for Aerospace Big Data [J]. Journal of Beijing University of Posts and Telecommunications, 2023, 46(3): 1-6.
[2]	OUYANG Mowei, KANG Guixia, WANG Kailiang, WANG Yaming. Automated Hippocampal Sclerosis Detection Method with Radiomics Analysis [J]. Journal of Beijing University of Posts and Telecommunications, 2022, 45(4): 51-57.
[3]	MU Xiao-hui, LI Li-xiang. Network Traffic Prediction of Dropout Echo State Network [J]. Journal of Beijing University of Posts and Telecommunications, 2021, 44(5): 10-13,20.
[4]	TANG Yong-li, LI Xing-yu, ZHAO Zong-qu, LI Yun-feng. Detection Method for Android Payment Cracked Application [J]. Journal of Beijing University of Posts and Telecommunications, 2021, 44(4): 95-101.
[5]	PENG Yu-he, CHEN Xiang, CHEN Shuang-wu, YANG Jian. Cross-Domain Abnormal Traffic Detection Based on Transfer Learning [J]. Journal of Beijing University of Posts and Telecommunications, 2021, 44(2): 33-39.
[6]	Lü Xing-feng, LI Jin-bao. A Method of Detecting Sleep Apnea Using Random Forest [J]. Journal of Beijing University of Posts and Telecommunications, 2020, 43(5): 64-70.
[7]	ZHANG De-gan, CHEN Lu, CHEN Chen, ZHANG Ting, CUI Yu-ya. A New Algorithm of QoS Constrained Routing for Node Energy Optimization of Edge Computing [J]. Journal of Beijing University of Posts and Telecommunications, 2020, 43(4): 101-105.
[8]	ZHOU Yi-lin, YANG Lu, LU Wen-rui. Exploring the Life Modeling Methods for Electrochemical Migration Failure of Printed Circuit Board under Dust Particles [J]. Journal of Beijing University of Posts and Telecommunications, 2020, 43(3): 11-18,31.
[9]	ZHANG Guo-dong, YING Huan, YANG Shou-guo, SHI Zhi-qiang, LI Ji-yuan. Automatic Identification and Cracking Method for Vulnerable Hash Functions of Embedded Firmwares [J]. Journal of Beijing University of Posts and Telecommunications, 2020, 43(1): 46-53.
[10]	YU Kai-xiang, CHEN Zhen-hao, ZHANG Si-hai. A Key Variable Detection Algorithm in Multivariate Manufacturing Process [J]. JOURNAL OF BEIJING UNIVERSITY OF POSTS AND TELECOM, 2019, 42(6): 98-104.
[11]	MENG Hao, HUO Ru, GUO Qian-ying, HUANG Tao, LIU Yun-jie. Machine Learning-Based Stochastic Task Offloading Algorithm in Mobile-Edge Computing [J]. JOURNAL OF BEIJING UNIVERSITY OF POSTS AND TELECOM, 2019, 42(2): 25-30.
[12]	CUI Hong-yan, XU Shuai, ZHANG Li-feng, Roy E. Welsch, Berthold K. P. Horn. The Key Techniques and Future Vision of Feature Selection in Machine Learning [J]. JOURNAL OF BEIJING UNIVERSITY OF POSTS AND TELECOM, 2018, 41(1): 1-12.
[13]	DONG Feng, LIU Tian-ming, XU Guo-ai, GUO Yan-hui, LI Cheng-ze. Defect Prediction Method for Android Binary Files [J]. JOURNAL OF BEIJING UNIVERSITY OF POSTS AND TELECOM, 2018, 41(1): 13-23.
[14]	ZENG Qing-tao, XIE Kai, LI Ye-li, WANG Xin-gang, YE Yu-shan, MA Shao-ping. Analysis Algorithm of Reference Record in HTML Page [J]. JOURNAL OF BEIJING UNIVERSITY OF POSTS AND TELECOM, 2017, 40(s1): 85-88.
[15]	Guochu Shou Yihong Hu. Network Flow Identification Based on Machine Learning [J]. JOURNAL OF BEIJING UNIVERSITY OF POSTS AND TELECOM, 2009, 32(1): 65-68.

Prosodic Structure Prediction based on Conditional Random Field Model

PDF

Knowledge

Abstract

Cite this article

share this article

References

Related Articles 15

Recommended Articles

Metrics

Comments