收稿日期: 2014-03-24
修回日期: 2014-05-16
网络出版日期: 2015-03-15
基金资助
国家自然科学基金(61005019,61273268,90920302)和北京市自然科学基金(KZ201110005005)资助
Design and implementation of mandarin spoken dialogue system for flight reservation
Received date: 2014-03-24
Revised date: 2014-05-16
Online published: 2015-03-15
介绍一个航班预定口语对话系统的设计与实现, 该系统允许用户通过普通话进行航班信息查询与预定.重点介绍口语对话系统中的口语语言理解.为了克服语音识别引入的识别错误导致语义理解错误的问题, 提出基于词混淆网络的两阶段中文口语语言理解方法:首先从词混淆网络中选择N元文法作为分类特征, 进行主题分类, 并通过语义分类模型解析获取对应的语义树结构;然后利用基于规则的语义槽填充器抽取相应的语义槽属性-值.该方法是数据驱动的, 训练数据的标记比较容易.实验在汉语航班预定领域进行, 结果表明, 在语音识别字错误率很高的情况下, 该方法比传统的基于语法规则的语言理解方法更加鲁棒, 在语义理解正确率方面有明显改善.
陈振锋 , 杨晓昊 , 吴蔚澜 , 刘加 , 夏善红 . 航班预定口语对话系统的设计与实现[J]. 中国科学院大学学报, 2015 , 32(2) : 252 -258 . DOI: 10.7523/j.issn.2095-6134.2015.02.015
We present a spoken dialogue system for flight reservation, which allows users to inquire information about flight in mandarin. We describe the design and the implementation of our system, focusing on spoken language understanding (SLU). Considering that the speech recognizer inevitably makes errors, we propose a new two-stage mandarin SLU approach based on word confusion network. Firstly, the semantic tuple classifier is used to identify the topic of an input utterance using N-gram features extracted from the word confusion network and to parse a semantic tree by recursively calling semantic classification models. Then the rule-based semantic slot filler is used to extract the corresponding slot/value pairs. The advantage of the proposed approach is that it is mainly data-driven and requires minimally annotated corpus for training. Experiment has been carried out in the Chinese flight reservation domain, which shows that the proposed approach is more robust to speech recognition errors than the conventional handcrafted rule-based parser, and substantially improves performance of accuracy when the ASR word error rate is high.
[1] Weng F, Cavedon L, Raghunathan B, et al. A conversational dialogue system for cognitively overloaded users [C]//International Conference on Spoken Language Processing. 2004.
[2] Liu J, Xu Y, Seneff S, et al. CityBrowser II: a multimodal restaurant guide in mandarin [C]//International Symposium on Chinese Spoken Language Processing. 2008: 1-4.
[3] Huang C, Xu P, Zhang X, et al. LODESTAR: a mandarin spoken dialogue system for travel information retrieval [C]//Proceedings of Eurospeech. 1999, 99: 1 159-1 162.
[4] 黄寅飞, 郑方,燕鹏举,等. 校园导航系统 EasyNav 的设计与实现[J]. 中文信息学报, 2001, 15(4): 35-40.
[5] ?ibert J, Martin ?i ?-Ipši ? S, Hajdinjak M, et al. Development of a bilingual spoken dialog system for weather information retrieval [C]//Proceedings of Eurospeech. 2003: 1 917-1 920.
[6] Lin Y C, Chiang T H, Wang H M, et al. The design of a multi-domain mandarin Chinese spoken dialogue system [C]//International Conference on Spoken Language Processing. 1998.
[7] Ward W H. The Phoenix system: understanding spontaneous speech [C]//Proceedings of ICASSP. 1991, 66.
[8] Ward W, Issar S. Recent improvements in the CMU spoken language understanding system [C]//Proceedings of the workshop on Human Language Technology. Association for Computational Linguistics, 1994: 213-216.
[9] Oerder M, Ney H. Word graphs: an efficient interface between continuous-speech recognition and language understanding [C]//Proceedings of ICASSP. 1993, 2: 119-122.
[10] Tür G, Wright J H, Gorin A L, et al. Improving spoken language understanding using word confusion networks [C]//Interspeech. 2002.
[11] Mangu L, Brill E, Stolcke A. Finding consensus among words: lattice-based word error minimization [C]//Proceedings of Eurospeech. 1999.
[12] Hakkani-Tür D, Béchet F, Riccardi G, et al. Beyond ASR 1-best: using word confusion networks in spoken language understanding[J]. Computer Speech & Language, 2006, 20(4): 495-514.
[13] Mairesse F, Gasic M, Jurcícek F, et al. Spoken language understanding from unaligned data using discriminative classification models [C]//Processing of ICASSP. 2009: 4 749-4 752.
[14] Chang C C, Lin C J. LIBSVM: a library for support vector machines[J]. ACM Transactions on Intelligent Systems and Technology (TIST), 2011, 2(3): 27.
[15] Lee C, Jung S, Kim K, et al. Recent approaches to dialog management for spoken dialog systems[J]. JCSE, 2010, 4(1): 1-22
[16] Cole R. Tools for research and education in speech science [C]//Proceedings of the International Conference of Phonetic Sciences. 1999: 1 277-1 280.
[17] Aust H, Schroer O. An overview of the Philips dialog system [C]//DARPA Broadcast News Transcription and Understanding Workshop, Lansdowne, VA. 1998.
[18] Bos J, Klein E, Lemon O, et al. DIPPER: Description and formalisation of an information-state update dialogue system architecture [C]//4th SIGdial Workshop on Discourse and Dialogue. 2003: 115-124.
[19] Ljunglöf P. trindikit. py: an open-source Python library for developing ISU-based dialogue systems[J]. Proceedings of IWSDS, 2009, 9.
[20] Rich C, Sidner C L. COLLAGEN: a collaboration manager for software interface agents[J]. User Modeling and User-Adapted Interaction, 1998, 8(3/4): 315-350.
[21] Bohus D, Rudnicky A I. The RavenClaw dialog management framework: architecture and systems[J]. Computer Speech & Language, 2009, 23(3): 332-361.
/
| 〈 |
|
〉 |
