DC Field | Value | Language |
---|---|---|
dc.contributor.author | Yamamoto, Yuya | ko |
dc.contributor.author | Nam, Juhan | ko |
dc.contributor.author | Terasawa, Hiroko | ko |
dc.date.accessioned | 2022-12-03T02:00:36Z | - |
dc.date.available | 2022-12-03T02:00:36Z | - |
dc.date.created | 2022-12-02 | - |
dc.date.created | 2022-12-02 | - |
dc.date.issued | 2022-09-21 | - |
dc.identifier.citation | 23rd Annual Conference of the International Speech Communication Association, INTERSPEECH 2022, pp.2778 - 2782 | - |
dc.identifier.issn | 2308-457X | - |
dc.identifier.uri | http://hdl.handle.net/10203/301509 | - |
dc.description.abstract | Singing techniques are used for expressive vocal performances by employing temporal fluctuations of the timbre, the pitch, and other components of the voice. Their classification is a challenging task, because of mainly two factors: 1) the fluctuations in singing techniques have a wide variety and are affected by many factors and 2) existing datasets are imbalanced. To deal with these problems, we developed a novel audio feature learning method based on deformable convolution with decoupled training of the feature extractor and the classifier using a class-weighted loss function. The experimental results show the following: 1) the deformable convolution improves the classification results, particularly when it is applied to the last two convolutional layers, and 2) both re-training the classifier and weighting the cross-entropy loss function by a smoothed inverse frequency enhance the classification performance. | - |
dc.language | English | - |
dc.publisher | International Speech Communication Association | - |
dc.title | Deformable CNN and Imbalance-Aware Feature Learning for Singing Technique Classification | - |
dc.type | Conference | - |
dc.identifier.wosid | 000900724502190 | - |
dc.identifier.scopusid | 2-s2.0-85140085822 | - |
dc.type.rims | CONF | - |
dc.citation.beginningpage | 2778 | - |
dc.citation.endingpage | 2782 | - |
dc.citation.publicationname | 23rd Annual Conference of the International Speech Communication Association, INTERSPEECH 2022 | - |
dc.identifier.conferencecountry | KO | - |
dc.identifier.conferencelocation | Incheon | - |
dc.identifier.doi | 10.21437/Interspeech.2022-11137 | - |
dc.contributor.localauthor | Nam, Juhan | - |
dc.contributor.nonIdAuthor | Yamamoto, Yuya | - |
dc.contributor.nonIdAuthor | Terasawa, Hiroko | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.