Presentation 1996/7/19
Detection of Ironical Utterance Basea on Acoustic Features
Hiroshi MITSUMOTO, Masuzo YANAGIDA, Hiroshi OTAWA, Shinichi TAMURA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Recent speech recognition research has been mainly focused on transcription or understanding of speech. A single sentence, however, can carry connotations or hidden meanings together with or besides its literal meaning in many cases. Speech recognition systems will be required to handle these cases in the future. In this paper, acoustic features of ironical utterance are first discussed and then a neural network approach for discriminating ironical utterances from literal ones is proposed with more than 70% discrimination score for speech data uttered by nine actors. Though the proposed approach is sentence dependent, the network seems to be able to detect possibility of hidden meanings in a given utterance leaving final decision for logical inference.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Irony / sensitivity / Prosody / Utterance recognition
Paper # SP96-39
Date of Issue

Conference Information
Committee SP
Conference Date 1996/7/19(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Speech (SP)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Detection of Ironical Utterance Basea on Acoustic Features
Sub Title (in English)
Keyword(1) Irony
Keyword(2) sensitivity
Keyword(3) Prosody
Keyword(4) Utterance recognition
1st Author's Name Hiroshi MITSUMOTO
1st Author's Affiliation Osaka Electro-Communication Junior College()
2nd Author's Name Masuzo YANAGIDA
2nd Author's Affiliation Faculty of Enginnering , Doshisha University
3rd Author's Name Hiroshi OTAWA
3rd Author's Affiliation Division of Functional Diagnostic Imaging, 0saka University Medical School
4th Author's Name Shinichi TAMURA
4th Author's Affiliation Division of Functional Diagnostic Imaging, 0saka University Medical School
Date 1996/7/19
Paper # SP96-39
Volume (vol) vol.96
Number (no) 160
Page pp.pp.-
#Pages 8
Date of Issue