Co-Operative Study on Modeling and Machine Inplementation of Spoken Language Conversation

口语对话建模与机器实现的合作研究

• 批准号: 02305010
• 负责人: SHIRAI Katsuhiko
• 金额: $ 4.67万
• 依托单位: Waseda University
• 依托单位国家: 日本
• 项目类别: Grant-in-Aid for Co-operative Research (A)
• 财政年份: 1990
• 资助国家: 日本
• 起止时间: 1990 至 1991
• 项目状态: 已结题
• 来源: https://kaken.nii.ac.jp/grant/KAKENHI-PROJECT-02305010/
• 关键词: Speech recognition.; Speech understanding.; Speech synthesis.; Natural language processing.; Conversation modeling.; 音声対話モデル; 連続音声認識; 対話デ-タベ-ス

Conversation using spoken language is the most fundamental forms in our communication. The modeling of the spoken language conversation is the main theme in various fields of study : linguistics, philosophy, cognitive science, etc. To realize a mechanism which performs the spoken language conversation in computer systems is indispensable to achieve advanced man-machine interface. The aims of our study are to analyze the spoken language conversation from the view points of behavioral science and linguistics, to model it using the method of information science, and to realize it on a computer system by the engineering technique.Ten meetings are held in these two years. Where, many researchers are participated not only from the inside of committee but also from the outside of it and enlivened the discussion.1. Language model for spoken conversation.2. Production of surface representation of spoken language in the spoken language conversation system.3. Understanding conversational speech.4. Mechanism of dialogue control.5. Database for spoken language conversation.One of the aims of our project is to start genuine research on spoken language conversation on the basis of the cooperation of speech researchers and language processing researchers. This aim is successfully achieved.

口语对话是人类交际中最基本的形式。口语会话的建模是语言学、哲学、认知科学等各个领域研究的主题。在计算机系统中实现口语会话的机制是实现高级人机接口的必要条件。我们的研究目的是从行为科学和语言学的角度分析口语会话，用信息科学的方法对其建模，并通过工程技术在计算机系统上实现。在那里，许多研究人员不仅从委员会内部参与，而且从委员会外部参与，活跃了讨论。口语对话的语言模型.口语会话系统中口语表层表征的产生.理解会话言语。4.对话控制机制。口语会话数据库。我们项目的目的之一是在语音研究者和语言处理研究者合作的基础上开始真正的口语会话研究。这一目标已成功实现。

项目成果

M.Nakatsui and R.Tanaka: "Half-syllabics for speech processing -An automatic segmentation" J.Acoust.Soc.Tpn (E). 11. 243-246 (1990)

M.Nakatsui 和 R.Tanaka：“语音处理的半音节 - 自动分段”J.Acoust.Soc.Tpn (E)。

S.Itahashi: "Creating Speech Corpora for Speech Science and Technology" IEICE Trans.E74. 1906-1910 (1991)

S.Itahashi：“为语音科学和技术创建语音语料库”IEICE Trans.E74。

小坂 直敏、筧 一彦: "Conversational TurnーTaking Model using Petri Net" ICSLP90. 1. 1297-1300 (1990)

Naotoshi Kosaka、Kazuhiko Kakei：“使用 Petri 网的对话轮转模型”ICSLP90。1. 1297-1300 (1990)。

榑松 明、武田 一哉、桑原 尚尺、鹿野 清宏: "ATR Japanese Speech Database as a tool of Speech Recognition and Synthesis" SPEECH COMMUNICATION. Vol.9No.4. 357-359 (1990)

Akira Kuramatsu、Kazuya Takeda、Naoshaku Kuwabara、Kiyohiro Kano：“ATR 日语语音数据库作为语音识别和合成的工具”SPEECH COMMUNICATION Vol.9No.4 (1990)。

森元 逞、飯田 仁、榑松 明、鹿野 清宏、相沢 輝昭: "Spoken language translation toward realizing an automatic telephone interpretation system" Proceeding of InfoーJapan'90(情報処理学会30周年記念国際会議). 553-560 (1990)

Takashi Morimoto、Hitoshi Iida、Akira Kuramatsu、Kiyohiro Kano、Teruaki Aizawa：“实现自动电话翻译系统的口语翻译”Proceeding of Info-Japan90（日本信息处理学会 30 周年国际会议）553-560。 1990））