Presentation 2002/10/23
Laughter Detection for VCML Player Caption Generation
Masafumi KURITA, Takahiro SUZUKI, Masahide SUGIYAMA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Generally speaking, captions in the broadcast and movies are designed for hearing-impaired people to acquire disaster information, business information and enjoy itself as amusement. However, researches on the conventional broadcasting with captions covers only linguistic contents, but do not cover auditory information. Hearing-impaired people strongly desire to get auditory information; laughter, applause, descriptive sounds, clapping, etc. This paper describes the VQ classifier-based laughter detection in speech data and the evaluation experiment experiments. The results of experiments show that the detection with time duration information gives high detection accuracy and the normalization technique of acoustic transfer characteristics improves the laughter detection accuracy for open data. The implementation of laughter display into VCML is described.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Video Caption / Laughter Detection / Auditory Scene / VQ Classifier / Duration
Paper # WIT2002-36
Date of Issue

Conference Information
Committee WIT
Conference Date 2002/10/23(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 Well-being Information Technology(WIT)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Laughter Detection for VCML Player Caption Generation
Sub Title (in English)
Keyword(1) Video Caption
Keyword(2) Laughter Detection
Keyword(3) Auditory Scene
Keyword(4) VQ Classifier
Keyword(5) Duration
1st Author's Name Masafumi KURITA
1st Author's Affiliation Graduate School of Computer Science and Engineering, The Univ. of Aizu()
2nd Author's Name Takahiro SUZUKI
2nd Author's Affiliation Graduate School of Computer Science and Engineering, The Univ. of Aizu
3rd Author's Name Masahide SUGIYAMA
3rd Author's Affiliation Graduate School of Computer Science and Engineering, The Univ. of Aizu
Date 2002/10/23
Paper # WIT2002-36
Volume (vol) vol.102
Number (no) 419
Page pp.pp.-
#Pages 6
Date of Issue