IEEE Kansai Section第6回技術講演会
IEEE Kansai Section第6回技術講演会
記
技術講演会概要
講演会テーマ(Title)
``The C/D Model: A Syllable-Based Representation of Conversational Speech''
講演者(Lecturers)
Prof. Osamu Fujimura
(The Ohio State University, Dept. Speech & Hearing Science)
日時(Time & Date)
2000年8月11日(金) 13:30～15:00
場所(Place)
ATR 地下大会議室(Conference Room)
〒619-0288 京都府相楽郡精華町光台 2-2-2
ATRの地図
講演内容（Abstract)
In our daily conversation, we use a set of words organized into phrases as a linguistic message. The utterance contains, however, additional information which is specified for a specific situation of verbal communication, to identify the speaker, convey the speaker's emotion, as well as intention and expectation. While such extra-linguistic information is crucial for effective conversation, speech technology has concentrated its effort on the capability of machines to produce and understand linguistic information, treating other phonetic information as both extraneous and disturbing factors. We need a new theory that can represent lively utterances as observed naturally. The Converter/Distributor (C/D) Model is a phonetic theory for representing realistic speech utterances. Words and phrases are represented abstractly as linear strings of syllables rather than consonants and vowels, and each syllable is specified as a set of features rather than a sequence of phonemic segments. Experimental observations of how conversational speech is produced as the articulatory processes of tongue and jaw movements have been interpreted by the Converter/Distributor (C/D) model, gradually revealing principles and strategies of speech organization. Widely variable utterance characteristics are now to be represented by an explicit computational process, making this abstract theory of speech experimentally testable. In order to achieve this goal, however, abstract representation theory of linguistic forms such as words and sentences also needs to be developed further, to elucidate basic characteristics of our cognitive process of speaking and hearing as well as writing and reading. Information technology cannot make a breakthrough for the new century without deeper understanding of language and speech.
講演者略歴(Biography)
Address: Department of Speech and Hearing Science, The Ohio State University, 1070 Carmack Rd., Columbus OH 43210-1002 Telephone/Fax/e-mail: 614 292 2769/7504/fujimura.1@osu.edu Birth: 29 August 1927, Tokyo (Citizenship: Japan. Permanent Residence: U. S. A.) Degrees: BS 1952, Faculty of Science (physics), University of Tokyo; D.Sc. 1962, Faculty of Science (physics), University of Tokyo Current position and year of appointment: Professor, 1988 (Dept. Speech & Hearing Science, Center for Cognitive Science, Biomedical Engineering Center) Previous positions in chronological order: 1958-1965 Assistant Professor, Univ. of Electrocommunications, Tokyo 1958-1961 DSR Research Staff Member, Research Laboratory of Electronics, MIT 1963-1965 Research Associate, Royal Institute of Technology, Stockholm 1965-1973 Professor, Faculty of Medicine (Research Institute of Logopedics and Phoniatrics) and Faculty of Letters (Dept. of Linguistics), University of Tokyo 1969-1973 Director, Research Institute of Logopedics and Phoniatrics 1973-1988 Head, AT&T Bell Laboratories (Depts. of Linguistics and Speech Analysis (73-84), Linguistics (84-87), Artificial Intelligence Research (87-88) Honors: The Asahi Award for Promotion of Science (1957) Shannon Award, NIH/NIDCD (1991) Acoustical Society of Japan Distinguished Contribution Award (1999) Advisory Review Panel, NSF Linguistics Program (1992-1995) NIH Expert Panels on Strategic Research Plan (1979, 1992) Executive Committee, Linguistic Society of America (1981-1983) Festschrift: S. Kiritani, H. Hirose & H. Fujisaki (eds.), Speech Production and Language: In Honor of Osamu Fujimura (Speech Research 13), Mouton de Gruyter (1997). Current professional activities: Editorial Boards: Phonetica, J. of Voice. Scientific Advisory Board: The Voice Foundation, New York; Organization Committee: LP Conference Series, Charles University, Prague. Fellow: Acoustical Society of America, The New York Academy of Sciences, American Association for the Advancement of Science. Sample Publications O. Fujimura, B. D. Joseph & B. Palek (eds.). Proceedings of LPユ98: Item Order in Language and Speech. Prague: Charles University Press (1999). O. Fujimura. The C/D model and prosodic control of articulatory behavior, Phonetica 2000 (in press). O. Fujimura. C/D model: A computational model of phonetic implementation. In S. Ristad (ed.), Language Computations, pp. 1-20. American Mathematical Society (1994). O. Fujimura. Methods and goals of speech production research, Language and Speech 33, 195-258 (1990).
参加費 (Fee)
無料 (Free）
参加申込み先 (Contact for registration)
619-288 京都府相楽郡精華町光台2-2
ATR知能映像通信研究所
Yumiko Honjo (ATR MIS Labs.)
Tel: (0774) 95 1404
Fax: (0774) 95 1408
E-mail: honjo@mic.atr.co.jp
会場準備の都合上、参加ご希望の方は、所属、お名前(ふりがな)、を上記までEmail（Faxでも結構です）にて8月4日（金）までにお知らせください。
本件連絡先 (For further information)
IEEE Kansai Section Technical Program Committee Secretary
間瀬 健二（ATR知能映像通信研究所）
E-mail: mase@mic.atr.co.jp
