AUTOENCODING HRTFS FOR DNN BASED HRTF PERSONALIZATION USING ANTHROPOMETRIC FEATURES
顯示簡易詮釋資料 統計資料| 欄位 | 內容 | 語言 |
|---|---|---|
dc.citation.epage |
275 | en_US |
dc.citation.spage |
271 | en_US |
dc.citation.woscount |
0 | en_US |
dc.contributor.author |
Chen, Tzu-Yu | en_US |
dc.contributor.author |
Kuo, Tzu-Hsuan | en_US |
dc.contributor.author |
Chi, Tai-Shih | en_US |
dc.contributor.department |
電機工程學系 | zh_TW |
dc.contributor.department |
Department of Electrical and Computer Engineering | en_US |
dc.date.accessioned |
2019-10-05T00:09:43Z | |
dc.date.available |
2019-10-05T00:09:43Z | |
dc.date.issued |
2019-01-01 | en_US |
dc.description.abstract |
We proposed a deep neural network (DNN) based approach to synthesize the magnitude of personalized head-related transfer functions (HRTFs) using anthropometric features of the user. To mitigate the over-fitting problem when training dataset is not very large, we built an autoencoder for dimensional reduction and establishing a crucial feature set to represent the raw HRTFs. Then we combined the decoder part of the autoencoder with a smaller DNN to synthesize the magnitude HRTFs. In this way, the complexity of the neural networks was greatly reduced to prevent unstable results with large variance due to overfitting. The proposed approach was compared with a baseline DNN model with no autoencoder. The log-spectral distortion (LSD) metric was used to evaluate the performance. Experiment results show that the proposed approach can reduce LSD of estimated HRTFs with greater stability. | en_US |
dc.identifier.isbn |
978-1-4799-8131-1 | en_US |
dc.identifier.issn |
1520-6149 | en_US |
dc.identifier.journal |
2019 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP) | en_US |
dc.identifier.uri |
https://ir.lib.nycu.edu.tw/handle/11536/152922 | |
dc.identifier.wosnumber |
WOS:000482554000055 | en_US |
dc.language.iso |
en_US | en_US |
dc.subject |
HRTFs | en_US |
dc.subject |
Anthropometry | en_US |
dc.subject |
Autoencoder | en_US |
dc.subject |
DNN | en_US |
dc.subject |
Spatial audio | en_US |
dc.title |
AUTOENCODING HRTFS FOR DNN BASED HRTF PERSONALIZATION USING ANTHROPOMETRIC FEATURES | en_US |
dc.type |
Proceedings Paper | en_US |
檔案
本條目沒有附加檔案。
所屬典藏: 會議論文
引用此條目
Chen, T.-Y., Kuo, T.-H., & Chi, T.-S. (2019). AUTOENCODING HRTFS FOR DNN BASED HRTF PERSONALIZATION USING ANTHROPOMETRIC FEATURES. 2019 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP), 271–275. https://ir.lib.nycu.edu.tw/handle/11536/152922
引用文字依 APA 第 7 版格式自動產生,使用前請再核對。
分享
永久網址:https://ir.lib.nycu.edu.tw/handle/11536/152922