| Presentation | 2023-03-03 Investigation of introducing data augmentation methods to improve speech enhancement performance Reito Kasuga, Yosuke Sugiura, Nozomiko Yasui, Tetsuya Shimamura, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | The field of speech enhancement has been extensively researched worldwide, and many speech enhancement methods have been proposed. However, there are still few large-scale speech datasets that can be used to train high-performance speech enhancement networks, making it difficult to create general-purpose models and making overlearning more likely to occur. In order to solve this problem, a method that enables the creation of large data sets from small data sets and improves the generalization performance of models is considered effective. In this paper, we attempt to solve the data size problem and improve the performance of speech enhancement by introducing a data expansion method called SpecAugment, which has demonstrated excellent performance in the field of speech recognition, to speech enhancement networks. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | speech enhancement / overlearning / speech recognition / SpecAugment / data augmentation |
| Paper # | SIS2022-52 |
| Date of Issue | 2023-02-23 (SIS) |
| Conference Information | |
| Committee | SIS |
|---|---|
| Conference Date | 2023/3/2(2days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | Chiba Institute of Technology |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | |
| Chair | Tomoaki Kimura(Kanagawa Inst. of Tech.) |
| Vice Chair | Naoto Sasaoka(Tottori Univ.) / Hakaru Tamukoh(Kyushu Inst. of Tech.) |
| Secretary | Naoto Sasaoka(NTT) / Hakaru Tamukoh(Kansai Univ.) |
| Assistant | Yoshiaki Makabe(Kanagawa Inst. of Tech.) / Yosuke Sugiura(Saitama Univ.) |
| Paper Information | |
| Registration To | Technical Committee on Smart Info-Media Systems |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Investigation of introducing data augmentation methods to improve speech enhancement performance |
| Sub Title (in English) | |
| Keyword(1) | speech enhancement |
| Keyword(2) | overlearning |
| Keyword(3) | speech recognition |
| Keyword(4) | SpecAugment |
| Keyword(5) | data augmentation |
| 1st Author's Name | Reito Kasuga |
| 1st Author's Affiliation | Saitama University(Saitama Univ.) |
| 2nd Author's Name | Yosuke Sugiura |
| 2nd Author's Affiliation | Saitama University(Saitama Univ.) |
| 3rd Author's Name | Nozomiko Yasui |
| 3rd Author's Affiliation | Saitama University(Saitama Univ.) |
| 4th Author's Name | Tetsuya Shimamura |
| 4th Author's Affiliation | Saitama University(Saitama Univ.) |
| Date | 2023-03-03 |
| Paper # | SIS2022-52 |
| Volume (vol) | vol.122 |
| Number (no) | SIS-410 |
| Page | pp.pp.64-69(SIS), |
| #Pages | 6 |
| Date of Issue | 2023-02-23 (SIS) |