Speech enhancement method based on multi-domain fusion and neural architecture search
In order to further improve the self-learning and noise reduction ability of speech enhancement model, a speech enhancement method based on multi-domain fusion and neural architecture search was proposed.The multi-spatial domain mapping and fusion mechanism of speech signals were designed to realize...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | zho |
| Published: |
Editorial Department of Journal on Communications
2024-02-01
|
| Series: | Tongxin xuebao |
| Subjects: | |
| Online Access: | http://www.joconline.com.cn/zh/article/doi/10.11959/j.issn.1000-436x.2024018/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841540030202380288 |
|---|---|
| author | Rui ZHANG Pengyun ZHANG Chaoli SUN |
| author_facet | Rui ZHANG Pengyun ZHANG Chaoli SUN |
| author_sort | Rui ZHANG |
| collection | DOAJ |
| description | In order to further improve the self-learning and noise reduction ability of speech enhancement model, a speech enhancement method based on multi-domain fusion and neural architecture search was proposed.The multi-spatial domain mapping and fusion mechanism of speech signals were designed to realize the mining of real complex number correlation.Based on the characteristics of convolution pooling of the model, a complex neural architecture search mechanism was proposed, and the speech enhancement model was constructed efficiently and automatically through the designed search space, search strategy and evaluation strategy.In the comparison and generalization experiment between the optimal speech enhancement model and the baseline model, the two indexes of PESQ and STOI increase by 5.6% compared with the optimal baseline model, and the number of model parameters is the lowest. |
| format | Article |
| id | doaj-art-32fd5f95ef0e42008ba209ec46b0665a |
| institution | Kabale University |
| issn | 1000-436X |
| language | zho |
| publishDate | 2024-02-01 |
| publisher | Editorial Department of Journal on Communications |
| record_format | Article |
| series | Tongxin xuebao |
| spelling | doaj-art-32fd5f95ef0e42008ba209ec46b0665a2025-01-14T06:22:10ZzhoEditorial Department of Journal on CommunicationsTongxin xuebao1000-436X2024-02-014522523959383535Speech enhancement method based on multi-domain fusion and neural architecture searchRui ZHANGPengyun ZHANGChaoli SUNIn order to further improve the self-learning and noise reduction ability of speech enhancement model, a speech enhancement method based on multi-domain fusion and neural architecture search was proposed.The multi-spatial domain mapping and fusion mechanism of speech signals were designed to realize the mining of real complex number correlation.Based on the characteristics of convolution pooling of the model, a complex neural architecture search mechanism was proposed, and the speech enhancement model was constructed efficiently and automatically through the designed search space, search strategy and evaluation strategy.In the comparison and generalization experiment between the optimal speech enhancement model and the baseline model, the two indexes of PESQ and STOI increase by 5.6% compared with the optimal baseline model, and the number of model parameters is the lowest.http://www.joconline.com.cn/zh/article/doi/10.11959/j.issn.1000-436x.2024018/speech enhancement modelcomplex spatial domain mappingmulti-domain fusioncomplex neural archi-tecture searchlow-cost evaluation |
| spellingShingle | Rui ZHANG Pengyun ZHANG Chaoli SUN Speech enhancement method based on multi-domain fusion and neural architecture search Tongxin xuebao speech enhancement model complex spatial domain mapping multi-domain fusion complex neural archi-tecture search low-cost evaluation |
| title | Speech enhancement method based on multi-domain fusion and neural architecture search |
| title_full | Speech enhancement method based on multi-domain fusion and neural architecture search |
| title_fullStr | Speech enhancement method based on multi-domain fusion and neural architecture search |
| title_full_unstemmed | Speech enhancement method based on multi-domain fusion and neural architecture search |
| title_short | Speech enhancement method based on multi-domain fusion and neural architecture search |
| title_sort | speech enhancement method based on multi domain fusion and neural architecture search |
| topic | speech enhancement model complex spatial domain mapping multi-domain fusion complex neural archi-tecture search low-cost evaluation |
| url | http://www.joconline.com.cn/zh/article/doi/10.11959/j.issn.1000-436x.2024018/ |
| work_keys_str_mv | AT ruizhang speechenhancementmethodbasedonmultidomainfusionandneuralarchitecturesearch AT pengyunzhang speechenhancementmethodbasedonmultidomainfusionandneuralarchitecturesearch AT chaolisun speechenhancementmethodbasedonmultidomainfusionandneuralarchitecturesearch |