| Presentation | 2022-01-20 [Short Paper] Estimation of Sound Environment in Room by applying to Transfer Learning using Spectrograms Images Shota Sano, Yuusuke Kawakita, Tsuyoshi Miyazaki, Hiroshi Tanaka, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | A variety of environmental sounds exist in a room, and the classification of these sounds can be expected to have various applications such as monitoring living conditions. The most common method for extracting environmental sounds is to remove unnecessary sounds as noise. However, in this experiment, the authors considered whether it is possible to estimate the entire sound environment by converting the acquired environmental sounds into spectrogram images and applying transfer learning. In the actual environment, the level of the environmental sound varies depending on the magnitude and distance of the sound source. Therefore, we created a learned model with multiple signal-to-noise ratios, and confirmed that the classification accuracy can be maintained even under these conditions. In addition, we verified the robustness of the learned model by performing classification on spectrogram images created under different conditions from those in the learned model. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Sound classification / Spectrogram / Transfer Learning / Indoor Sound Environment / Robustness |
| Paper # | SeMI2021-57 |
| Date of Issue | 2022-01-13 (SeMI) |
| Conference Information | |
| Committee | SeMI |
|---|---|
| Conference Date | 2022/1/20(2days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | |
| Chair | Koji Yamamoto(Kyoto Univ.) |
| Vice Chair | Kazuya Monden(Hitachi) / Yasunori Owada(NICT) |
| Secretary | Kazuya Monden(Cyber Univ.) / Yasunori Owada(Waseda Univ.) |
| Assistant | Yuki Katsumata(NTT DOCOMO) / Akihito Taya(Aoyama Gakuin Univ.) / Yu Nakayama(Tokyo Univ. of Agri. and Tech.) |
| Paper Information | |
| Registration To | Technical Committee on Sensor Network and Mobile Intelligence |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | [Short Paper] Estimation of Sound Environment in Room by applying to Transfer Learning using Spectrograms Images |
| Sub Title (in English) | |
| Keyword(1) | Sound classification |
| Keyword(2) | Spectrogram |
| Keyword(3) | Transfer Learning |
| Keyword(4) | Indoor Sound Environment |
| Keyword(5) | Robustness |
| 1st Author's Name | Shota Sano |
| 1st Author's Affiliation | Kanagawa Institute of Technology(KAIT) |
| 2nd Author's Name | Yuusuke Kawakita |
| 2nd Author's Affiliation | Kanagawa Institute of Technology(KAIT) |
| 3rd Author's Name | Tsuyoshi Miyazaki |
| 3rd Author's Affiliation | Kanagawa Institute of Technology(KAIT) |
| 4th Author's Name | Hiroshi Tanaka |
| 4th Author's Affiliation | Kanagawa Institute of Technology(KAIT) |
| Date | 2022-01-20 |
| Paper # | SeMI2021-57 |
| Volume (vol) | vol.121 |
| Number (no) | SeMI-333 |
| Page | pp.pp.22-25(SeMI), |
| #Pages | 4 |
| Date of Issue | 2022-01-13 (SeMI) |