| Presentation | 2021-07-16 Sound event detection based on complementary-label learning Keigo Wakayama, Shoichiro Saito, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | Sound Event Detection (SED) is an important research field that can be applied to smart cities, and etc. SED estimate the start and end of events in addition to the presence or absence of events in acoustic data. To improve the estimation accuracy, a lot of labeled data is required, but it is difficult to assign a lot of correct labels. Considering effectively utilizing the data with the complementary label (class that is not the correct answer), therefore, we propose a method to realize SED by learning model from complementary labeled data. The effectiveness of the method is confirmed experimentally. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Sound Event Detection / Audio Tagging / Weakly-supervised learning / Complementary-label learning |
| Paper # | EA2021-17 |
| Date of Issue | 2021-07-08 (EA) |
| Conference Information | |
| Committee | EA / ASJ-H |
|---|---|
| Conference Date | 2021/7/15(2days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | Online |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | Engineering/Electro Acoustics, Psychological and Physiological Acoustics, Speech, Musical Acoustics, Education in Acoustics, and Related Topics |
| Chair | Yoshinobu Kajikawa(Kansai Univ.) |
| Vice Chair | Kenichi Furuya(Oita Univ.) / Shoichi Koyama(Univ. of Tokyo) |
| Secretary | Kenichi Furuya(NTT) / Shoichi Koyama(RitsumeikanUniv.) |
| Assistant | Yukou Wakabayashi(Tokyo Metropolitan Univ.) / Tatsuya Komatsu(LINE) |
| Paper Information | |
| Registration To | Technical Committee on Engineering Acoustics / Auditory Research Meeting |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Sound event detection based on complementary-label learning |
| Sub Title (in English) | |
| Keyword(1) | Sound Event Detection |
| Keyword(2) | Audio Tagging |
| Keyword(3) | Weakly-supervised learning |
| Keyword(4) | Complementary-label learning |
| 1st Author's Name | Keigo Wakayama |
| 1st Author's Affiliation | Nippon Telegraph and Telephone Corporation(NTT) |
| 2nd Author's Name | Shoichiro Saito |
| 2nd Author's Affiliation | Nippon Telegraph and Telephone Corporation(NTT) |
| Date | 2021-07-16 |
| Paper # | EA2021-17 |
| Volume (vol) | vol.121 |
| Number (no) | EA-112 |
| Page | pp.pp.77-82(EA), |
| #Pages | 6 |
| Date of Issue | 2021-07-08 (EA) |