研究领域
智能语音信息处理,机器学习,包括:
(1)语音识别:大词汇量连续语音识别声学建模,声学模型自适应,语音唤醒和识别技术等
(2)声纹识别:Anti-spoofingspeakerverification,end-to-endspeakerrecognition等
(3)语音信号处理:远场语音信号的噪声鲁棒性研究,如语音增强,麦克风阵列信号处理等
科研获奖及学术评测:
1.2018年,“上海产学研合作优秀项目奖”二等奖,第一完成人。
2.2016年,国际“中英文混合语音识别竞赛(OC16Chinese-EnglishMixASRChallenge)”,提交的“SHNU”中英文混合语音识别系统取得国际第2名。
3.2018年7-8月,第5届国际多通道语音分离和识别评测(The5thCHiMESpeechSeparationandRecognitionChallenge,CHiME-5),提交的“SHNU系统”成绩排在国际第9名。
4.2019年1-2月,国际“防攻击声纹识别评测(AutomaticSpeakerVerificationSpoofingandCountermeasuresChallenge,ASVspoof2019)”,提交的“SHNU系统”成绩排在国际第13/156名。
5.2019年9月,联合实验室学生参加“多通道远场文本相关声纹识别-AISHELLSpeakerVerificationChallenge2019”竞赛,获得第4名/50.
6.2008NISTSpeakerRecognitionEvaluation(SRE),在核心测试任务中,作为关键技术人员及组长带领的团队获得EER、minDCF两项国际第一名,DCF第三名,综合成绩国际第一,该成果被国家自然科学基金委,中国科学院网站等100多家媒体报导。
7.2009NISTLanguageRecognitionEvaluation,团队在通用语种测试中各项指标综合排名国际第二;同时,在更具挑战性的8组方言对测试中,有6组方言对测试性能均远远超过了其他参赛单位,综合排名国际第一。
8.2010NISTSpeakerRecognitionEvaluation,作为关键技术人员及组长带领的团队获得EER,minDCF,DCF指标综合成绩国际第二名
主持的科研项目:
1.校一般科研项目,面向语音识别的副语言信息标注算法研究,已结题。.
2.上海市青年科技英才扬帆计划,基于深度学习的声纹识别方法研究,已结题.
3.企业横向课题-联盟计划项目,多语种混合语音识别开发,已结题.
4.校产学研项目,噪声环境下中英混合的语音识别系统研发,在研.
5.国家自然科学基金项目,中英文混合语音识别中声学建模关键技术研究,在研.
6.联合实验室横向课题,自然人机交互关键技术研发,在研.
7.2018联盟计划项目,防攻击的声纹识别关键技术研发,在研
近期论文
查看导师最新文章
(温馨提示:请注意重名现象,建议点开原文通过作者单位确认)
[1].Y.Long,Q.Zhang,S.Wei,H.Ye,J.Yang.AcousticdataaugmentationforMandarin-Englishcode-switchingspeechrecognition,2019,AppliedAcoustics,online:https://doi.org/10.1016/j.apacoust.2019.107175.
[2].R.He,Y.Long,Y.Li,J.Liang.Mask-basedblindsourceseparationandMVDRbeamforminginASR,InternationalJournalofSpeechTechnology,2019,online:http://link.springer.com/article/10.1007/s10772-019-09666-x
[3].Y.Shi,J.Zhou,Y.Long,Y.Li,H.Mao.AddressingText-DependentSpeakerVerificationUsingSingingSpeech.AppliedSciences,2019,9(13),2636.
[4].Y.Long,S.Wei,Q.Zhang,C.Yang.Large-ScaleSemi-SupervisedTraininginDeepLearningAcousticModelforASR.IEEEAccess,2019,(7):133615-133627.
[5].Z.Feng,Q.Tong,Y.Long,S.Weiandet.al.SHNUAnti-spoofingsystemsforASVspoof2019Challenge,APSIPA2019,pp.548-552.
[6].Y.Long,Y.Li,B.Zhang.Offlinetoonlinespeakeradaptationforreal-timedeepneuralnetworkbasedLVCSRsystems.MultimediaToolsandApplications,2018,77(21):28101-28119.
[7].Y.Long,R.He.TheSHNUSystemfortheCHiME-5Challenge.Proc.CHiME2018WorkshoponSpeechProcessinginEverydayEnvironments.2018:64-66.
[8].YanhuaLong,HongYe,YijieLi,JiaenLiang.ActiveLearningforLF-MMITrainedNeuralNetworksinASR.Proc.Interspeech,2018:2898-2902.
[9].YanZhang,YanhuaLong,XiangrongShen,et.al.Articulatorymovementfeaturesforshort-durationtext-dependentspeakerverification.InternationalJournalofSpeechTechnology,2017,20(4):753-759.
[10].YanhuaLong,YijieLi,HongYe.HongweiMao.Domainadaptationoflattice-freeMMIbasedTDNNmodelsforspeechrecognition.InternationalJournalofSpeechTechnology,2017,20(1):171-178.
[11].YanhuaLong,et.al.DomainCompensationBasedonPhoneticallyDiscriminativeFeaturesforSpeakerVerification,ComputerSpeech&Language,2017,(41):161-179.
[12].HaoranWei,YanhuaLong,et.al.Improvementsonself-adaptivevoiceactivitydetectorfortelephonedata,InternationalJournalofSpeechTechnology,2016,19(3):623-630.
[13].龙艳花,倪继锋,叶宏.“基于深度神经网络的说话人信道自适应方法”,2016,48(2):151-155.
[14].YanhuaLong,HongYe.FilledPauseRefinementBasedonthePronunciationProbabilityforLectureSpeech,PLosOne,10(4):2015,e0123466.doi:10.1371/journal.pone.0123466.
[15].BoLi,YanhuaLong,HongYe.OutlierDetectionandClusterCenterInitializationforK-meansAlgorithm,JournalofComputationalInformationSystems,11(12):2015,4333–4342.
[16].龙艳花,戴礼荣.“采用M-矢量和支持向量机的说话人确认系统”.华中科技大学学报(自然科学版),2014,42(8):63-68.
[17].Y.Long,M.J.F.Gales,P.Lanchantin,X.Liu,M.S.Seigel,P.C.Woodland.“ImprovingLightlySupervisedTrainingforBroadcastTranscription”.Interspeech,pp.2187-2191,2013.
[18].P.Lanchantin,P.Bell,M.J.F.Gales,T.Hain,X.Liu,Y.Long,J.Quinnell,S.Renals,O.Saz,M.S.Seigel,P.Swietojanski,P.C.Woodland.“Automatictranscriptionofmulti-genremediaarchives”.SLAM,pp.26-31,2013.
[19].P.Bell,M.Gales,P.Lanchantin,X.Liu,Y.Long,S.Renals,P.Swietojanski,P.C.Woodland,“TranscriptionofMulti-GenreMediaArchivesUsingOut-of-domaindata”,SLT,pp.324-329,2012.
[20].YanhuaLong,Zhi-JieYan,FrankKSoong,et.al.“ImprovementsinSpeakerCharacterizationUsingSpectralSubbandEnergyBasedonHarmonicplusNoiseModel”,pp.373-376,INTERSPEECH,2011.
[21].YanhuaLong,Zhi-JieYan,FrankKSoong,et.al.“SpeakerCharacterizationusingSpectralSubbandEnergyRatiobasedonHarmonicPlusNoiseModel”,pp.4520-4523,ICASSP,2011.
[22].YingXU,YanSong,Yan-HuaLong,et.al.”TheDescriptionofiFlyTekSpeechLabSystemforNIST2009LanguageRecognitionEvaluation”,pp.157-161,ISCSLP,2010.
[23].YanhuaLong,LiRongDai,Er-yuWang,et.al.“Non-negativematrixfactorizationbaseddiscriminativefeaturesforspeakerverification”,pp.291-295,ISCSLP,2010.
[24].WuGuo,YanhuaLong,EryuWang,er.al.“IFlyspeechlab2010speakerrecognitionevaluationsystemdescription”.NISTSRE2010,systemdescriptionpaper.(NISTSRE2010Evaluationpaper)
[25].YanhuaLong,LiRongDai,BinMa,WuGuo.“EffectsofthePhonologicalRelevanceinSpeakerVerification”,pp.2130-2133,INTERSPEECH,2010.
[26].WuGuo,ZhaoZhang,YanhuaLong,LirongDai.“N-gramNearestNeighborAlgorithmforVoicePasswordSystem”,pp.4438-4441,ICASSP,2010.
[27].YanhuaLong,BinMa,HaizhouLi,et.al.“ExploitingProsodicInformationforSpeakerRecognition”,pp.4225-4228,ICASSP,2009.
[28].WuGuo,YanhuaLong,YijieLi,et.al.“iFLYsystemfortheNIST2008speakerrecognitionevaluation”,pp.4209–4212,ICASSP,2009.
[29].YanhuaLong,WuGuo,BinMa,et.al.“SubspaceConstructionandSelectionforSpeakerRecognition”,pp.1-4,ICICS,2009.
[30].YanhuaLong,WuGuo,Lirongdai.“APCAMethodBasedonSpeakerSessionVariability”,JournalofPatternrecognitionandartificialintelligence,pp.270-274,No.22,Issue2,2009.
[31].YanhuaLong,WuGuo,LiRongDai.“ToBalanceTrainingDataforSVMBasedSpeakerVerification“,JournalofChineseInformationProcessing,pp.76-80,No.5,Issue3,2008.(Chinesecorejournals)
[32].YanhuaLong,WuGuo,LiRongDai.”AnSIPCA-WCCNMethodforSVM-basedSpeakerVerificationSystem”,pp.1295–1299,ICALIP,2008
[33].YanhuaLong,WuGuo,LiRongDai.”InterfusingtheConfusedRegionScoreofSpeakerVerificationSystems”,pp.1-4,ISCSLP2008.
[34].YanhuaLong,WuGuo,LiRongDai.”SequenceKernelforSVMbasedSpeakerverificationsystem”,JournalofTsinghuaUniversity(ScienceandTechnology),pp.688-692,Vol.48,No.S1,2008.