{"title":"Multi-channel speech enhancement in driving environment","authors":"Weiyun Jin, Jie Wei, Xiaofeng Zhong","doi":"10.1109/ICSPCS.2017.8270478","DOIUrl":null,"url":null,"abstract":"Speech enhancement algorithms based on microphone array have been applied to intelligent speech communication systems, such as vehicle systems, teleconferencing and hearing aids. This paper proposed an improved generalized sidelobe canceller(GSC) algorithm to eliminate the intense noise interference of the speech, which is recorded in the driving environment. First, it introduced low-frequency suppression; then, it utilized source location to improve the blocking matrix, thus effectively reducing the speech leakage; finally, the multi-band spectral subtraction algorithm was combined for denoising model. Experimental results show that the proposed microphone array based speech enhancement method has a significant performance in terms of both SNR and intelligibility.","PeriodicalId":268205,"journal":{"name":"2017 11th International Conference on Signal Processing and Communication Systems (ICSPCS)","volume":"64 6","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2017-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2017 11th International Conference on Signal Processing and Communication Systems (ICSPCS)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICSPCS.2017.8270478","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
Speech enhancement algorithms based on microphone array have been applied to intelligent speech communication systems, such as vehicle systems, teleconferencing and hearing aids. This paper proposed an improved generalized sidelobe canceller(GSC) algorithm to eliminate the intense noise interference of the speech, which is recorded in the driving environment. First, it introduced low-frequency suppression; then, it utilized source location to improve the blocking matrix, thus effectively reducing the speech leakage; finally, the multi-band spectral subtraction algorithm was combined for denoising model. Experimental results show that the proposed microphone array based speech enhancement method has a significant performance in terms of both SNR and intelligibility.