A screening scheme based on energy for speech key-frame

Yingjie Meng, Xiaoyu Teng, Huiran Liu, Sanshuai Cui, Zhiyuan Wang
{"title":"A screening scheme based on energy for speech key-frame","authors":"Yingjie Meng, Xiaoyu Teng, Huiran Liu, Sanshuai Cui, Zhiyuan Wang","doi":"10.1109/ICCT.2017.8359908","DOIUrl":null,"url":null,"abstract":"The research of the existing screening algorithm for speech frame has a great deal of shortcomings, such as its applicability and complexity. Worse still, those frames which are screened by algorithm can't achieve the requirement of express and the screening process greatly damages the original signal. This paper presents a strategy for screening speech keyframe and designs a screening scheme for key-frames based on the strategy. This scheme refers to the speech's logarithm energy and the weighted-zero-crossing rate. The detail process of screening scheme: firstly, screening frames according to the logarithm energy of the speech signals. Meanwhile, combine the speech amplitude and zero-crossing rate for frames screening. Finally, calculate the similarity of the two screening results, and getting the key-frame set. In addition, the scheme has been analyzed and validated from those aspects like continuity, characterization and applicability, in order to verify the effectiveness and availability. The results illustrate that these frames which have been screened by this scheme have advantages of continuity, characterization etc.","PeriodicalId":199874,"journal":{"name":"2017 IEEE 17th International Conference on Communication Technology (ICCT)","volume":"30 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2017 IEEE 17th International Conference on Communication Technology (ICCT)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICCT.2017.8359908","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 1

Abstract

The research of the existing screening algorithm for speech frame has a great deal of shortcomings, such as its applicability and complexity. Worse still, those frames which are screened by algorithm can't achieve the requirement of express and the screening process greatly damages the original signal. This paper presents a strategy for screening speech keyframe and designs a screening scheme for key-frames based on the strategy. This scheme refers to the speech's logarithm energy and the weighted-zero-crossing rate. The detail process of screening scheme: firstly, screening frames according to the logarithm energy of the speech signals. Meanwhile, combine the speech amplitude and zero-crossing rate for frames screening. Finally, calculate the similarity of the two screening results, and getting the key-frame set. In addition, the scheme has been analyzed and validated from those aspects like continuity, characterization and applicability, in order to verify the effectiveness and availability. The results illustrate that these frames which have been screened by this scheme have advantages of continuity, characterization etc.
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
基于能量的语音关键帧筛选方案
现有的语音帧筛选算法的研究存在着适用性和复杂性等诸多不足。更糟糕的是,算法筛选出来的帧不能达到表达的要求,而且筛选过程对原始信号的破坏很大。提出了一种语音关键帧的筛选策略,并在此基础上设计了一种关键帧的筛选方案。该方案是指语音的对数能量和加权过零率。筛选方案的详细过程:首先,根据语音信号的对数能量进行帧筛选。同时,结合语音幅值和过零率进行帧筛选。最后,计算两次筛选结果的相似度,得到关键帧集。并从连续性、表征性、适用性等方面对方案进行了分析和验证,验证了方案的有效性和可用性。结果表明,用该方案筛选出来的帧具有连续性、特征化等优点。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 去求助
来源期刊
自引率
0.00%
发文量
0
期刊最新文献
Chemical substance classification using long short-term memory recurrent neural network One-way time transfer for large area through tropospheric scatter Application feature extraction by using both dynamic binary tracking and statistical learning Research on multi-target resolution process with the same beam of monopulse radar Pedestrian detection based on Visconti2 7502
×
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