An improved deep learning approach for speech enhancement

Q4 Engineering
Malek Miled, M. B. Ben Messaoud
{"title":"An improved deep learning approach for speech enhancement","authors":"Malek Miled, M. B. Ben Messaoud","doi":"10.24840/2183-6493_009-005_001531","DOIUrl":null,"url":null,"abstract":"Single-channel speech enhancement refers to the task of improving the quality and intelligibility of a speech signal in a noisy environment. Time-domain and time-frequency-domain methods are two main categories of approaches for speech enhancement. In this paper, we propose a approach based on a cross-domain framework. This framework utilizes our knowledge of the spectrogram and overcomes some of the limitations faced by time-frequency domain methods. First, we apply the intrinsic mode functions of the empirical mode decomposition and an improved version of principal component analysis. Then, we design a cross-domain learning framework to determine the correlations along the frequency and time axes. At low SNR = -5 dB, the effectiveness of our proposed approach is demonstrated by its performance based on objective and subjective measures. With average scores of -0.49, 2.47, 2.44, and 0.68 for SegSNR, PESQ, Cov, and STOI, respectively. The results highlight the success of our approach in addressing low SNR conditions.","PeriodicalId":36339,"journal":{"name":"U.Porto Journal of Engineering","volume":"10 1","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2023-11-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"U.Porto Journal of Engineering","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.24840/2183-6493_009-005_001531","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"Engineering","Score":null,"Total":0}
引用次数: 0

Abstract

Single-channel speech enhancement refers to the task of improving the quality and intelligibility of a speech signal in a noisy environment. Time-domain and time-frequency-domain methods are two main categories of approaches for speech enhancement. In this paper, we propose a approach based on a cross-domain framework. This framework utilizes our knowledge of the spectrogram and overcomes some of the limitations faced by time-frequency domain methods. First, we apply the intrinsic mode functions of the empirical mode decomposition and an improved version of principal component analysis. Then, we design a cross-domain learning framework to determine the correlations along the frequency and time axes. At low SNR = -5 dB, the effectiveness of our proposed approach is demonstrated by its performance based on objective and subjective measures. With average scores of -0.49, 2.47, 2.44, and 0.68 for SegSNR, PESQ, Cov, and STOI, respectively. The results highlight the success of our approach in addressing low SNR conditions.
用于语音增强的改进型深度学习方法
单通道语音增强是指在噪声环境中提高语音信号的质量和可懂度。时域和时频域方法是语音增强的两大类方法。在本文中,我们提出了一种基于跨域框架的方法。该框架利用了我们对频谱图的了解,克服了时频域方法所面临的一些限制。首先,我们应用了经验模式分解的固有模式函数和改进版的主成分分析。然后,我们设计了一个跨域学习框架,以确定沿频率轴和时间轴的相关性。在 SNR = -5 dB 的低信噪比条件下,我们提出的方法的有效性通过其基于客观和主观测量的性能得到了证明。SegSNR、PESQ、Cov 和 STOI 的平均得分分别为 -0.49、2.47、2.44 和 0.68。这些结果凸显了我们的方法在解决低信噪比条件下的成功。
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 求助全文
来源期刊
U.Porto Journal of Engineering
U.Porto Journal of Engineering Engineering-Engineering (all)
CiteScore
0.70
自引率
0.00%
发文量
58
审稿时长
20 weeks
×
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