Presentation | 2014-01-24 Automatic Recognition of Paralinguistic Information with Speaker Dependent Modeling Tomoyuki SHIMAKAWA, Yoichi YAMASHITA, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | This paper discusses about the automatic recognition of paralinguistic information with speaker dependent modeling. When a user uses a spoken dialog system which can use paralinguistic information, it requires high performance of the paralinguistic information recognition for the specific user. We tried to recognize paralinguistic information in a speaker dependent manner. We recorded speech with paralinguistic information, and tried automatic recognition of three kinds of paralinguistic information, dissatisfaction, interrogation, and surprise, which can be perceived in a listening test. The average recognition rate is about 70% by a method based on weighted summation of output from several recognizers which use a single parameter. We tried some conditions of model training, and compared the recognition results with speaker dependent and speaker independent modeling. We could find the difference of recognition for speakers, but we didn't find the larger difference between the speaker dependent and speaker independent modeling. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | Paralinguistic information / Spoken dialog / Emotion / Speaker Dependent |
Paper # | SP2013-103 |
Date of Issue |
Conference Information | |
Committee | SP |
---|---|
Conference Date | 2014/1/16(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) | Automatic Recognition of Paralinguistic Information with Speaker Dependent Modeling |
Sub Title (in English) | |
Keyword(1) | Paralinguistic information |
Keyword(2) | Spoken dialog |
Keyword(3) | Emotion |
Keyword(4) | Speaker Dependent |
1st Author's Name | Tomoyuki SHIMAKAWA |
1st Author's Affiliation | Graduate school of Information Science and Engineering, Ritsumeikan University() |
2nd Author's Name | Yoichi YAMASHITA |
2nd Author's Affiliation | Information Science and Engineering, Ritsumeikan University |
Date | 2014-01-24 |
Paper # | SP2013-103 |
Volume (vol) | vol.113 |
Number (no) | 404 |
Page | pp.pp.- |
#Pages | 6 |
Date of Issue |