Prediction of Speech Pauses Based on Punctuation Information and Statistical Language Model
QIAN Yi-Li1,2, XUN En-Dong3
1.College of Computer Science, Beijing University of Technology, Beijing 1000222. College of Computer and Information Technology, Shanxi University, Taiyuan 0300063. College of Information Sciences, Beijing Language and Culture University, Beijing 100083
Abstract Speech pauses are considered as punctuation marks of spoken language. People always insert different pauses at the boundaries of rhythmic phrases when communicating by language. Based on this characteristic, the speech pause of punctuation marks is investigated and the concept of predicting speech pauses using punctuation information is proposed. The punctuation-based and SLM-based methods are introduced to obtain training corpus and predict speech pauses. The influence of training corpus size on the performance of model is discussed. And the performance of punctuation-based corpus and manually-labeled corpus is compared. Experimental results show that the Chinese punctuation supplies valuable information on pause, and the method based on punctuation information can predict the Chinese speech pauses effectively.
[1] Zheng Min, Cai Lianhong. Statistical Model Based on Probability Frequency for Mandarin Prosodic Structure Prediction. Journal of Tsinghua University: Science and Technology, 2006, 46(1): 78-81 (in Chinese) (郑 敏,蔡莲红.基于概率频度的普通话韵律结构预测统计模型.清华大学学报:自然科学版, 2006, 46(1): 78-81) [2] Li Jianfeng, Hu Guoping, Wang Renhua. Prosody Phrase Break Prediction Based on Maximum Entropy Model. Journal of Chinese Information Processing, 2004, 18(5): 56-63 (in Chinese) (李剑锋,胡国平,王仁华.基于最大熵模型的韵律短语边界预测.中文信息学报, 2004,18(5): 56-63) [3] Cao Jianfen. Prediction of Prosodic Organization Based on Grammatical Information. Journal of Chinese Information Processing, 2003, 17(3): 41-46 (in Chinese) (曹剑芬.基于语法信息的汉语韵律结构预测.中文信息学报, 2003, 17(3): 41-46) [4] Zhao Sheng, Tao Jianhua, Cai Lianhong. Rule-Learning Based Prosodic Structure Prediction. Journal of Chinese Information Processing, 2002, 16(5): 30-37 (in Chinese) (赵 晟,陶建华,蔡莲红.基于规则学习的韵律结构预测.中文信息学报, 2002,16(5): 30-37) [5] Niu Zhengyu, Chai Peiqi. A Statistical Approach Based on Boundary POS Feature to Prosodic Phrasing. Journal of Chinese Information Processing, 2001, 15(5): 19-25 (in Chinese) (牛正雨, 柴佩琪. 基于边界点词性特征统计的韵律短语切分. 中文信息学报, 2001,15(5): 19-25) [6] Ying Hong, Cai Lianhong. Research on the Segmentation of the Prosodic Phrase Based on Driven by the Structural Auxiliary Word. Journal of Chinese Information Processing, 1999, 13(6): 41-46 (in Chinese) (应 宏,蔡莲红.基于结构助词驱动韵律短语界定的研究.中文信息学报, 1999, 13(6): 41-46) [7] Nie Xin, Wang Zuoying. Automatic Phrase Breaks Prediction in Chinese Sentences. Journal of Chinese Information Processing, 2003, 17(4): 39-44 (in Chinese) (聂 鑫,王作英.汉语语句中短语间停顿的自动预测方法. 中文信息学报, 2003,17(4): 39-44) [8] Chu Min, Yao Qian. Locating Boundaries for Prosodic Constituents in Unrestricted Mandarin Texts. Computational Linguistics and Chinese Language Processing, 2001, 6(1): 61-82 [9] Yang Jinchen, Yang Yufang. Prosody Generation in Language Production. Advances in Psychological Science, 2004, 12(4): 481-488 (in Chinese) (杨锦陈, 杨玉芳. 言语产生中的韵律生成. 心理科学进展, 2004,12(4): 481-488) [10] Ostendorf M, Veilleux N. A Hierarchical Stochastic Model for Automatic Prediction of Prosodic Boundary Location. Computational Linguistics, 1994, 20(1): 27-54
2008, Vol. 21 
