Few-shot short utterance speaker verification using meta-learning
Short utterance speaker verification (SV) in the actual application is the task of accepting or rejecting the identity claim of a speaker based on a few enrollment utterances. Traditional methods have used deep neural networks to extract speaker representations for verification. Recently, several me...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
PeerJ Inc.
2023-04-01
|
| Series: | PeerJ Computer Science |
| Subjects: | |
| Online Access: | https://peerj.com/articles/cs-1276.pdf |
| _version_ | 1797840755669073920 |
|---|---|
| author | Weijie Wang Hong Zhao Yikun Yang YouKang Chang Haojie You |
| author_facet | Weijie Wang Hong Zhao Yikun Yang YouKang Chang Haojie You |
| author_sort | Weijie Wang |
| collection | DOAJ |
| description | Short utterance speaker verification (SV) in the actual application is the task of accepting or rejecting the identity claim of a speaker based on a few enrollment utterances. Traditional methods have used deep neural networks to extract speaker representations for verification. Recently, several meta-learning approaches have learned a deep distance metric to distinguish speakers within meta-tasks. Among them, a prototypical network learns a metric space that may be used to compute the distance to the prototype center of speakers, in order to classify speaker identity. We use emphasized channel attention, propagation and aggregation in TDNN (ECAPA-TDNN) to implement the necessary function for the prototypical network, which is a nonlinear mapping from the input space to the metric space for either few-shot SV task. In addition, optimizing only for speakers in given meta-tasks cannot be sufficient to learn distinctive speaker features. Thus, we used an episodic training strategy, in which the classes of the support and query sets correspond to the classes of the entire training set, further improving the model performance. The proposed model outperforms comparison models on the VoxCeleb1 dataset and has a wide range of practical applications. |
| first_indexed | 2024-04-09T16:19:51Z |
| format | Article |
| id | doaj.art-e8c5426d5a28405c89ed4ffda40c8d30 |
| institution | Directory Open Access Journal |
| issn | 2376-5992 |
| language | English |
| last_indexed | 2024-04-09T16:19:51Z |
| publishDate | 2023-04-01 |
| publisher | PeerJ Inc. |
| record_format | Article |
| series | PeerJ Computer Science |
| spelling | doaj.art-e8c5426d5a28405c89ed4ffda40c8d302023-04-23T15:05:06ZengPeerJ Inc.PeerJ Computer Science2376-59922023-04-019e127610.7717/peerj-cs.1276Few-shot short utterance speaker verification using meta-learningWeijie Wang0Hong Zhao1Yikun Yang2YouKang Chang3Haojie You4School of Computer and Communication, Lanzhou University of Technology, Lanzhou, ChinaSchool of Computer and Communication, Lanzhou University of Technology, Lanzhou, ChinaSchool of Information Science & Engineering, Lanzhou University, Lanzhou, ChinaSchool of Computer and Communication, Lanzhou University of Technology, Lanzhou, ChinaSchool of Computer and Communication, Lanzhou University of Technology, Lanzhou, ChinaShort utterance speaker verification (SV) in the actual application is the task of accepting or rejecting the identity claim of a speaker based on a few enrollment utterances. Traditional methods have used deep neural networks to extract speaker representations for verification. Recently, several meta-learning approaches have learned a deep distance metric to distinguish speakers within meta-tasks. Among them, a prototypical network learns a metric space that may be used to compute the distance to the prototype center of speakers, in order to classify speaker identity. We use emphasized channel attention, propagation and aggregation in TDNN (ECAPA-TDNN) to implement the necessary function for the prototypical network, which is a nonlinear mapping from the input space to the metric space for either few-shot SV task. In addition, optimizing only for speakers in given meta-tasks cannot be sufficient to learn distinctive speaker features. Thus, we used an episodic training strategy, in which the classes of the support and query sets correspond to the classes of the entire training set, further improving the model performance. The proposed model outperforms comparison models on the VoxCeleb1 dataset and has a wide range of practical applications.https://peerj.com/articles/cs-1276.pdfSpeaker verificationMeta-learningSupport setPrototypical networkGlobal classificationEpisodic training strategy |
| spellingShingle | Weijie Wang Hong Zhao Yikun Yang YouKang Chang Haojie You Few-shot short utterance speaker verification using meta-learning PeerJ Computer Science Speaker verification Meta-learning Support set Prototypical network Global classification Episodic training strategy |
| title | Few-shot short utterance speaker verification using meta-learning |
| title_full | Few-shot short utterance speaker verification using meta-learning |
| title_fullStr | Few-shot short utterance speaker verification using meta-learning |
| title_full_unstemmed | Few-shot short utterance speaker verification using meta-learning |
| title_short | Few-shot short utterance speaker verification using meta-learning |
| title_sort | few shot short utterance speaker verification using meta learning |
| topic | Speaker verification Meta-learning Support set Prototypical network Global classification Episodic training strategy |
| url | https://peerj.com/articles/cs-1276.pdf |
| work_keys_str_mv | AT weijiewang fewshotshortutterancespeakerverificationusingmetalearning AT hongzhao fewshotshortutterancespeakerverificationusingmetalearning AT yikunyang fewshotshortutterancespeakerverificationusingmetalearning AT youkangchang fewshotshortutterancespeakerverificationusingmetalearning AT haojieyou fewshotshortutterancespeakerverificationusingmetalearning |