深度神经网络中用于文本无关说话人验证的分段单元混洗层

Ju-Sung Heo, Hye-jin Shim, Ju-ho Kim, Ha-jin Yu
{"title":"深度神经网络中用于文本无关说话人验证的分段单元混洗层","authors":"Ju-Sung Heo, Hye-jin Shim, Ju-ho Kim, Ha-jin Yu","doi":"10.7776/ASK.2021.40.2.148","DOIUrl":null,"url":null,"abstract":"Text-Independent speaker verification needs to extract text-independent speaker embedding to improve generalization performance. However, deep neural networks that depend on training data have the potential to overfit text information instead of learning the speaker information when repeatedly learning from the identical time series. In this paper, to prevent the overfitting, we propose a segment unit shuffling layer that divides and rearranges the input layer or a hidden layer along the time axis, thus mixes the time series information. Since the segment unit shuffling layer can be applied not only to the input layer but also to the hidden layers, it can be used as generalization technique in the hidden layer, which is known to be effective compared to the generalization technique in the input layer, and can be applied simultaneously with data augmentation. In addition, the degree of distortion can be adjusted by adjusting the unit size of the segment. We observe that the performance of text-independent speaker verification is improved compared to the baseline when the proposed segment unit shuffling layer is applied.","PeriodicalId":42689,"journal":{"name":"Journal of the Acoustical Society of Korea","volume":"40 1","pages":"148-154"},"PeriodicalIF":0.2000,"publicationDate":"2021-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Segment unit shuffling layer in deep neural networks for text-independent speaker verification\",\"authors\":\"Ju-Sung Heo, Hye-jin Shim, Ju-ho Kim, Ha-jin Yu\",\"doi\":\"10.7776/ASK.2021.40.2.148\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Text-Independent speaker verification needs to extract text-independent speaker embedding to improve generalization performance. However, deep neural networks that depend on training data have the potential to overfit text information instead of learning the speaker information when repeatedly learning from the identical time series. In this paper, to prevent the overfitting, we propose a segment unit shuffling layer that divides and rearranges the input layer or a hidden layer along the time axis, thus mixes the time series information. Since the segment unit shuffling layer can be applied not only to the input layer but also to the hidden layers, it can be used as generalization technique in the hidden layer, which is known to be effective compared to the generalization technique in the input layer, and can be applied simultaneously with data augmentation. In addition, the degree of distortion can be adjusted by adjusting the unit size of the segment. We observe that the performance of text-independent speaker verification is improved compared to the baseline when the proposed segment unit shuffling layer is applied.\",\"PeriodicalId\":42689,\"journal\":{\"name\":\"Journal of the Acoustical Society of Korea\",\"volume\":\"40 1\",\"pages\":\"148-154\"},\"PeriodicalIF\":0.2000,\"publicationDate\":\"2021-03-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Journal of the Acoustical Society of Korea\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.7776/ASK.2021.40.2.148\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q4\",\"JCRName\":\"ACOUSTICS\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of the Acoustical Society of Korea","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.7776/ASK.2021.40.2.148","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"ACOUSTICS","Score":null,"Total":0}
引用次数: 0

摘要

文本无关说话人验证需要提取文本无关说话人嵌入,以提高泛化性能。然而,当从相同的时间序列中重复学习时,依赖于训练数据的深度神经网络有可能过度拟合文本信息,而不是学习说话者信息。在本文中,为了防止过拟合,我们提出了一个分段单元混洗层,它沿着时间轴划分和重新排列输入层或隐藏层,从而混合时间序列信息。由于分段单元混洗层不仅可以应用于输入层,还可以应用于隐藏层,因此它可以用作隐藏层中的泛化技术,与输入层中的推广技术相比,它是有效的,并且可以与数据扩充同时应用。此外,可以通过调整片段的单位大小来调整失真程度。我们观察到,当应用所提出的分段单元混洗层时,与基线相比,与文本无关的说话人验证的性能有所提高。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
Segment unit shuffling layer in deep neural networks for text-independent speaker verification
Text-Independent speaker verification needs to extract text-independent speaker embedding to improve generalization performance. However, deep neural networks that depend on training data have the potential to overfit text information instead of learning the speaker information when repeatedly learning from the identical time series. In this paper, to prevent the overfitting, we propose a segment unit shuffling layer that divides and rearranges the input layer or a hidden layer along the time axis, thus mixes the time series information. Since the segment unit shuffling layer can be applied not only to the input layer but also to the hidden layers, it can be used as generalization technique in the hidden layer, which is known to be effective compared to the generalization technique in the input layer, and can be applied simultaneously with data augmentation. In addition, the degree of distortion can be adjusted by adjusting the unit size of the segment. We observe that the performance of text-independent speaker verification is improved compared to the baseline when the proposed segment unit shuffling layer is applied.
求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
CiteScore
0.60
自引率
50.00%
发文量
1
期刊最新文献
A quantitative analysis of synthetic aperture sonar image distortion according to sonar platform motion parameters Measurements of mid-frequency transmission loss in shallow waters off the East Sea: Comparison with Rayleigh reflection model and high-frequency bottom loss model An explorative study on the perceived emotion of music: according to cognitive styles of music listening A robust data association gate method of non-linear target tracking in dense cluttered environment Performance analysis of weakly-supervised sound event detection system based on the mean-teacher convolutional recurrent neural network model
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1