DC Field | Value | Language |
---|---|---|
dc.contributor.author | Jung, Jee-weon | ko |
dc.contributor.author | Kim, Youjin | ko |
dc.contributor.author | Heo, Hee-Soo | ko |
dc.contributor.author | Lee, Bong-Jin | ko |
dc.contributor.author | Kwon, Youngki | ko |
dc.contributor.author | Chung, Joon Son | ko |
dc.date.accessioned | 2023-05-10T12:01:23Z | - |
dc.date.available | 2023-05-10T12:01:23Z | - |
dc.date.created | 2023-05-03 | - |
dc.date.issued | 2022-09 | - |
dc.identifier.citation | 23rd Annual Conference of the International Speech Communication Association, INTERSPEECH 2022, pp.2228 - 2232 | - |
dc.identifier.issn | 2308-457X | - |
dc.identifier.uri | http://hdl.handle.net/10203/306697 | - |
dc.description.abstract | In recent years, speaker recognition systems based on raw waveform inputs have received increasing attention. However, the performance of such systems are typically inferior to the state-of-the-art handcrafted feature-based counterparts, which demonstrate equal error rates under 1% on the popular VoxCeleb1 test set. This paper proposes a novel speaker recognition model based on raw waveform inputs. The model incorporates recent advances in machine learning and speaker verification, including the Res2Net backbone module and multi-layer feature aggregation. Our best model achieves an equal error rate of 0.89%, which is competitive with the state-of-the-art models based on handcrafted features, and outperforms the best model based on raw waveform inputs by a large margin. We also explore the application of the proposed model in the context of self-supervised learning framework. Our self-supervised model outperforms single phase-based existing works in this line of research. Finally, we show that self-supervised pre-training is effective for the semi-supervised scenario where we only have a small set of labelled training data, along with a larger set of unlabelled examples. | - |
dc.language | English | - |
dc.publisher | ISCA-INT SPEECH COMMUNICATION ASSOC | - |
dc.title | Pushing the limits of raw waveform speaker recognition | - |
dc.type | Conference | - |
dc.identifier.wosid | 000900724502081 | - |
dc.identifier.scopusid | 2-s2.0-85140069445 | - |
dc.type.rims | CONF | - |
dc.citation.beginningpage | 2228 | - |
dc.citation.endingpage | 2232 | - |
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.2022-126 | - |
dc.contributor.localauthor | Chung, Joon Son | - |
dc.contributor.nonIdAuthor | Jung, Jee-weon | - |
dc.contributor.nonIdAuthor | Kim, Youjin | - |
dc.contributor.nonIdAuthor | Heo, Hee-Soo | - |
dc.contributor.nonIdAuthor | Lee, Bong-Jin | - |
dc.contributor.nonIdAuthor | Kwon, Youngki | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.