分布移位下的唤醒词检测

S. Parthasarathi, Lu Zeng, Christin Jose, Joe Wang
{"title":"分布移位下的唤醒词检测","authors":"S. Parthasarathi, Lu Zeng, Christin Jose, Joe Wang","doi":"10.48550/arXiv.2207.06423","DOIUrl":null,"url":null,"abstract":"We propose a novel approach for semi-supervised learning (SSL) designed to overcome distribution shifts between training and real-world data arising in the keyword spotting (KWS) task. Shifts from training data distribution are a key challenge for real-world KWS tasks: when a new model is deployed on device, the gating of the accepted data undergoes a shift in distribution, making the problem of timely updates via subsequent deployments hard. Despite the shift, we assume that the marginal distributions on labels do not change. We utilize a modified teacher/student training framework, where labeled training data is augmented with unlabeled data. Note that the teacher does not have access to the new distribution as well. To train effectively with a mix of human and teacher labeled data, we develop a teacher labeling strategy based on confidence heuristics to reduce entropy on the label distribution from the teacher model; the data is then sampled to match the marginal distribution on the labels. Large scale experimental results show that a convolutional neural network (CNN) trained on far-field audio, and evaluated on far-field audio drawn from a different distribution, obtains a 14.3% relative improvement in false discovery rate (FDR) at equal false reject rate (FRR), while yielding a 5% improvement in FDR under no distribution shift. Under a more severe distribution shift from far-field to near-field audio with a smaller fully connected network (FCN) our approach achieves a 52% relative improvement in FDR at equal FRR, while yielding a 20% relative improvement in FDR on the original distribution.","PeriodicalId":358274,"journal":{"name":"International Conference on Text, Speech and Dialogue","volume":"19 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2022-07-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":"{\"title\":\"Wakeword Detection under Distribution Shifts\",\"authors\":\"S. Parthasarathi, Lu Zeng, Christin Jose, Joe Wang\",\"doi\":\"10.48550/arXiv.2207.06423\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"We propose a novel approach for semi-supervised learning (SSL) designed to overcome distribution shifts between training and real-world data arising in the keyword spotting (KWS) task. Shifts from training data distribution are a key challenge for real-world KWS tasks: when a new model is deployed on device, the gating of the accepted data undergoes a shift in distribution, making the problem of timely updates via subsequent deployments hard. Despite the shift, we assume that the marginal distributions on labels do not change. We utilize a modified teacher/student training framework, where labeled training data is augmented with unlabeled data. Note that the teacher does not have access to the new distribution as well. To train effectively with a mix of human and teacher labeled data, we develop a teacher labeling strategy based on confidence heuristics to reduce entropy on the label distribution from the teacher model; the data is then sampled to match the marginal distribution on the labels. Large scale experimental results show that a convolutional neural network (CNN) trained on far-field audio, and evaluated on far-field audio drawn from a different distribution, obtains a 14.3% relative improvement in false discovery rate (FDR) at equal false reject rate (FRR), while yielding a 5% improvement in FDR under no distribution shift. Under a more severe distribution shift from far-field to near-field audio with a smaller fully connected network (FCN) our approach achieves a 52% relative improvement in FDR at equal FRR, while yielding a 20% relative improvement in FDR on the original distribution.\",\"PeriodicalId\":358274,\"journal\":{\"name\":\"International Conference on Text, Speech and Dialogue\",\"volume\":\"19 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2022-07-13\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"1\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"International Conference on Text, Speech and Dialogue\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.48550/arXiv.2207.06423\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"International Conference on Text, Speech and Dialogue","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.48550/arXiv.2207.06423","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 1

摘要

我们提出了一种新的半监督学习(SSL)方法,旨在克服关键字识别(KWS)任务中出现的训练数据和真实数据之间的分布变化。训练数据分布的变化是现实世界KWS任务的一个关键挑战:当在设备上部署新模型时,可接受数据的门接在分布上发生变化,使得通过后续部署及时更新的问题变得困难。尽管有这种变化,我们假设标签上的边际分布不变。我们使用修改后的教师/学生培训框架,其中标记的培训数据与未标记的数据相增强。请注意,老师也没有访问新分发的权限。为了有效地训练人类和教师标记数据的混合,我们开发了一种基于置信度启发式的教师标记策略,以减少来自教师模型的标签分布的熵;然后对数据进行采样以匹配标签上的边际分布。大规模实验结果表明,卷积神经网络(CNN)在远场音频上进行训练,并对来自不同分布的远场音频进行评估,在相同的错误拒绝率(FRR)下,错误发现率(FDR)相对提高14.3%,而在没有分布移动的情况下,FDR提高了5%。在从远场到近场音频的更严重的分布转移和更小的完全连接网络(FCN)下,我们的方法在相同的FRR下实现了52%的FDR相对改进,同时在原始分布上产生了20%的FDR相对改进。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
Wakeword Detection under Distribution Shifts
We propose a novel approach for semi-supervised learning (SSL) designed to overcome distribution shifts between training and real-world data arising in the keyword spotting (KWS) task. Shifts from training data distribution are a key challenge for real-world KWS tasks: when a new model is deployed on device, the gating of the accepted data undergoes a shift in distribution, making the problem of timely updates via subsequent deployments hard. Despite the shift, we assume that the marginal distributions on labels do not change. We utilize a modified teacher/student training framework, where labeled training data is augmented with unlabeled data. Note that the teacher does not have access to the new distribution as well. To train effectively with a mix of human and teacher labeled data, we develop a teacher labeling strategy based on confidence heuristics to reduce entropy on the label distribution from the teacher model; the data is then sampled to match the marginal distribution on the labels. Large scale experimental results show that a convolutional neural network (CNN) trained on far-field audio, and evaluated on far-field audio drawn from a different distribution, obtains a 14.3% relative improvement in false discovery rate (FDR) at equal false reject rate (FRR), while yielding a 5% improvement in FDR under no distribution shift. Under a more severe distribution shift from far-field to near-field audio with a smaller fully connected network (FCN) our approach achieves a 52% relative improvement in FDR at equal FRR, while yielding a 20% relative improvement in FDR on the original distribution.
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
自引率
0.00%
发文量
0
×
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
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
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 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信