| Presentation | 2018-11-21 Evaluation of DNN-based Low-Musical-Noise Speech Enhancement Using Kurtosis Matching Satoshi Mizoguchi, Yuki Saito, Shinnosuke Takamichi, Hiroshi Saruwatari, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | This paper proposes DNN-based speech enhancement with low musical noise by kurtosis matching. Musical noise, artifacts generated by nonlinear signal processing, causes a negative effect on the auditory impression. Quantity of the generated musical noise is significantly correlated with increase in kurtosis from observed signal to enhanced signal. Although soft-mask-based DNN speech enhancement has a high performance on noise reduction thanks to rich power of expression of DNN, it does not consider generation of musical noise. This paper proposes low-musical-noise speech enhancement without degrading noise-reduction-rate and generating significant speech distortion by applying kurtosis matching, which is regularization to prevent kurtosis from increasing, to DNN-based speech enhancement. We give objective evaluation of the enhanced speech signal to demonstrate the efficiency of the proposed method. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | speech enhancement / musical noise / kurtosis matching / deep learning |
| Paper # | EA2018-66,EMM2018-66 |
| Date of Issue | 2018-11-14 (EA, EMM) |
| Conference Information | |
| Committee | EA / ASJ-H / EMM / IPSJ-MUS |
|---|---|
| Conference Date | 2018/11/21(2days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | Hotel Koshuen |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | [Beginners Session] Engineering/Electro Acoustics, Psychological and Physiological Acoustics, Music and Computer, Content Processing, Digital Watermarking, and Related Topics |
| Chair | Suehiro Shimauchi(Kanazawa Inst. of Tech.) / 小澤 賢司(山梨大) / Keiichi Iwamura(TUC) / 吉井 和佳(京大) |
| Vice Chair | Kenichi Furuya(Oita Univ.) / Kanji Watanabe(Akita Pref. Univ.) / 中川 誠司(千葉大) / Minoru Kuribayashi(Okayama Univ.) / Tetsuya Kojima(NIT,Tokyo College) |
| Secretary | Kenichi Furuya(Shizuoka Inst. of Science and Tech.) / Kanji Watanabe(NHK) / 中川 誠司(尚絅大) / Minoru Kuribayashi(北陸先端大) / Tetsuya Kojima(千葉大) / (NIT, Tokyo) |
| Assistant | Keisuke Imoto(Ritsumeikan Univ.) / Daisuke Morikawa(Toyama Pref Univ.) / / Hiroko Akiyama(NIT, Nagano College) / キタヒロ カネダ(CANON) |
| Paper Information | |
| Registration To | Technical Committee on Engineering Acoustics / Auditory Research Meeting / Technical Committee on Enriched MultiMedia / Special Interest Group on Music and Computer |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Evaluation of DNN-based Low-Musical-Noise Speech Enhancement Using Kurtosis Matching |
| Sub Title (in English) | |
| Keyword(1) | speech enhancement |
| Keyword(2) | musical noise |
| Keyword(3) | kurtosis matching |
| Keyword(4) | deep learning |
| 1st Author's Name | Satoshi Mizoguchi |
| 1st Author's Affiliation | The University of Tokyo(UTokyo) |
| 2nd Author's Name | Yuki Saito |
| 2nd Author's Affiliation | The University of Tokyo(UTokyo) |
| 3rd Author's Name | Shinnosuke Takamichi |
| 3rd Author's Affiliation | The University of Tokyo(UTokyo) |
| 4th Author's Name | Hiroshi Saruwatari |
| 4th Author's Affiliation | The University of Tokyo(UTokyo) |
| Date | 2018-11-21 |
| Paper # | EA2018-66,EMM2018-66 |
| Volume (vol) | vol.118 |
| Number (no) | EA-312,EMM-313 |
| Page | pp.pp.19-24(EA), pp.19-24(EMM), |
| #Pages | 6 |
| Date of Issue | 2018-11-14 (EA, EMM) |