Full metadata record
DC FieldValueLanguage
dc.contributor.authorChen, Tzu-Yuen_US
dc.contributor.authorKuo, Tzu-Hsuanen_US
dc.contributor.authorChi, Tai-Shihen_US
dc.date.accessioned2019-10-05T00:09:43Z-
dc.date.available2019-10-05T00:09:43Z-
dc.date.issued2019-01-01en_US
dc.identifier.isbn978-1-4799-8131-1en_US
dc.identifier.issn1520-6149en_US
dc.identifier.urihttp://hdl.handle.net/11536/152922-
dc.description.abstractWe 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.language.isoen_USen_US
dc.subjectHRTFsen_US
dc.subjectAnthropometryen_US
dc.subjectAutoencoderen_US
dc.subjectDNNen_US
dc.subjectSpatial audioen_US
dc.titleAUTOENCODING HRTFS FOR DNN BASED HRTF PERSONALIZATION USING ANTHROPOMETRIC FEATURESen_US
dc.typeProceedings Paperen_US
dc.identifier.journal2019 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP)en_US
dc.citation.spage271en_US
dc.citation.epage275en_US
dc.contributor.department電機工程學系zh_TW
dc.contributor.departmentDepartment of Electrical and Computer Engineeringen_US
dc.identifier.wosnumberWOS:000482554000055en_US
dc.citation.woscount0en_US
Appears in Collections:Conferences Paper