| Presentation | 2005/5/19 A Study on Feature Values relating to Emotion Katsuya NAOI, Tetsuya MATSUMOTO, Yoshinori TAKEUCHI, Hiroaki KUDO, Noboru OHNISHI, |
|---|---|
| PDF Download Page | PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | This paper investigates feature values relating to emotion in speech. In conventional studies on recognition of speech emotion, a pitch and energy have been used as feature values. There is, however, problem that these features depend on the length of attended words. Then we aim for finding features of speech emotion which don't depend on words. We propose a new feature value of peaks in frequency domain. We conducted two experiments: two-emotion discrimination and four-emotion discrimination by using four emotions of anger, joy, sadness and normal. From the result of two-emotion discrimination, it is said that the proposed feature does not discriminate between normal and sadness, but does joy and sadness. In the experiment of four emotions, the average discrimination rate is 58%, and the rate for joy is 77%. Although the learning was done without considering the length of speech, there is no change in the discrimination rate. This result shows that the proposed feature doesn't depend on spoken words. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | signal / emotion recognition / frequency domain / Teager Energy Operator |
| Paper # | HIP2005-7 |
| Date of Issue |
| Conference Information | |
| Committee | HIP |
|---|---|
| Conference Date | 2005/5/19(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 | Human Information Processing (HIP) |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | A Study on Feature Values relating to Emotion |
| Sub Title (in English) | |
| Keyword(1) | signal |
| Keyword(2) | emotion recognition |
| Keyword(3) | frequency domain |
| Keyword(4) | Teager Energy Operator |
| 1st Author's Name | Katsuya NAOI |
| 1st Author's Affiliation | Graduate school of Information Science, Nagoya University() |
| 2nd Author's Name | Tetsuya MATSUMOTO |
| 2nd Author's Affiliation | Graduate school of Information Science, Nagoya University |
| 3rd Author's Name | Yoshinori TAKEUCHI |
| 3rd Author's Affiliation | Information Security Promotion Agency, Nagoya University; BMC Research Center, RIKEN |
| 4th Author's Name | Hiroaki KUDO |
| 4th Author's Affiliation | Graduate school of Information Science, Nagoya University |
| 5th Author's Name | Noboru OHNISHI |
| 5th Author's Affiliation | Graduate school of Information Science, Nagoya University |
| Date | 2005/5/19 |
| Paper # | HIP2005-7 |
| Volume (vol) | vol.105 |
| Number (no) | 99 |
| Page | pp.pp.- |
| #Pages | 6 |
| Date of Issue | |