Article(id=1329420246915592311, tenantId=1146029695717560320, journalId=1324759914146529478, issueId=1329420206000157303, articleNumber=null, orderNo=null, doi=10.7544/issn1000-1239.202550673, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=null, receivedDate=1757433600000, receivedDateStr=2025-09-10, revisedDate=1773244800000, revisedDateStr=2026-03-12, acceptedDate=1775145600000, acceptedDateStr=2026-04-03, onlineDate=1794888086297, onlineDateStr=2026-11-17, pubDate=1785513600000, pubDateStr=2026-08-01, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1794888086297, onlineIssueDateStr=2026-11-17, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1794888086297, creator=13701087609, updateTime=1794888086297, updator=13701087609, issue=Issue{id=1329420206000157303, tenantId=1146029695717560320, journalId=1324759914146529478, year='2026', volume='63', issue='8', pageStart='1905', pageEnd='2181', issueExtLink='null', onlineDate='null', pubDate='1785513600000', pubDateStr='2026-08-01', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=-1, specialIssue=null, createTime=1794888076542, creator='13701087609', updateTime=1794888076542, updator='13701087609', preIssue=null, nextIssue=null, articleTotal=null, ext=null, issueFiles=null, downloadFileDto=null}, startPage=2095, endPage=2110, ext={EN=ArticleExt(id=1329420247104335992, articleId=1329420246915592311, tenantId=1146029695717560320, journalId=1324759914146529478, language=EN, title=Voice Conversion Spoofing Dataset and Model for Source Speaker Verification, columnId=1325489891789910943, journalTitle=Journal of Computer Research and Development, columnName=Artificial Intelligence, runingTitle=null, highlight=null, articleAbstract=

With the rapid development of deep learning-based voice conversion (VC) technology, attackers can compromise the security of automatic speaker verification (ASV) systems by tampering with the source speaker’s speech signal to mimic the target speaker’s acoustic characteristics. The existing anti-spoofing countermeasures primarily focus on distinguishing between genuine and spoofed speech, yet underutilize the implicit source speaker identity information concealed within the spoofed speech. This limitation hinders their ability to meet practical demands in scenarios such as forensic investigation and fraud traceability. Currently, research on source speaker identity verification and discrimination remains insufficient, and existing methods suffer from limited generalisation capabilities. To address this, we construct a new voice conversion spoofing dataset specifically designed for source speaker verification, named VCSD-SSV (voice conversion spoofing dataset for source speaker verification). This dataset encompasses six VC algorithms (i.e., Seed-VC, FACodec, DDDM-VC, Diff-HierVC, TriAAN-VC, and Free-VC) and includes 400 speakers, totaling 327 677 speech samples (46 811 genuine samples and 280 866 converted samples). Building upon this, we propose a feature generalisation enhancement-based source speaker verification model, FGE-SSV (source speaker verification based on feature generalised enhancement). This model extracts multi-level source speaker timbre features through a feature enhancement module and cascaded instance-batch normalization squeeze-and-excitation blocks. Experimental results demonstrate that FGE-SSV maintains consistently low equal error rate (EER) (0.2%~1.0%) in within-domain results across all six VC methods in VCSD-SSV dataset and also achieves relatively low EER across the six VC methods in SSTC2024 dataset. Furthermore, FGE-SSV demonstrates strong generalization performance on unseen voice conversion methods, with its EER values outperforming those of comparative methods.

, authors=Guofu Zhang1, 2, 3, Tiancai Tian1, Zhaopin Su1, 3, *, Yaofei Wang1, 3, Yuheng Duan1, Kangjian Zhou1, Yongqi Luo1, authorsList=Guofu Zhang, Tiancai Tian, Zhaopin Su, Yaofei Wang, Yuheng Duan, Kangjian Zhou, Yongqi Luo, authorCompany=null, correspAuthors=Zhaopin Su, authorNote=

born in 1979. PhD, professor. His main research interests include artificial intelligence and audio security

born in 2001. Master candidate. His main research interest includes deepfake audio detection and source tracing

born in 1983. PhD, associate professor. Her main research interest includes network and information security

born in 1996. PhD, associate professor. His main research interests include information security and artificial intelligence security

born in 2004. Undergraduate. His main research interest includes network and information security

born in 2005. Undergraduate. His main research interest includes network and information security

born in 2005. Undergraduate. His main research interest includes network and information security

, correspAuthorsNote=null, copyrightStatement=Copyright © 2026 Journal of Computer Research and Development. All rights reserved., copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=null, magXml=null, pdfUrl=null, pdf=null, pdfFileSize=null, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=null, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=null, mapNumber=null, fund=null), CN=ArticleExt(id=1329420250350727326, articleId=1329420246915592311, tenantId=1146029695717560320, journalId=1324759914146529478, language=CN, title=面向源说话人验证的语音转换伪造数据集与模型, columnId=1325489892104483745, journalTitle=计算机研究与发展, columnName=人工智能, runingTitle=null, highlight=null, articleAbstract=

随着基于深度学习的语音转换(voice conversion,VC)技术的快速发展,攻击者能够通过篡改源说话人的语音信号来模仿目标说话人的声学特征,这严重威胁自动说话人验证(automatic speaker verification,ASV)系统的安全性。现有反欺骗防御机制主要聚焦于区分真实语音和伪造语音,却未能充分利用伪造语音中隐含的源说话人身份信息,导致其在司法取证、欺诈溯源等实际应用场景中难以满足需求。目前,针对源说话人身份验证与辨别的研究尚不充分,且已有模型泛化性能有限。基于此,构建了一个专门用于源说话人验证的新语音转换伪造数据集(voice conversion spoofing dataset for source speaker verification,VCSD-SSV)。该数据集涵盖Seed-VC,FACodec,DDDM-VC,Diff-HierVC,TriAAN-VC,Free-VC共6种语音转换算法,包含400个说话人,总计327 677条语音样本(其中真实语音46 811条、转换语音280 866条)。在此基础上,设计了一种基于特征泛化增强的源说话人验证(source speaker verification based on feature generalised enhancement,FGE-SSV)模型。该模型通过特征增强模块和级联的实例-批量归一化挤压激励块实现多层次源说话人音色特征的提取。实验结果表明,FGE-SSV在VCSD-SSV数据集全部6种VC模型中域内结果保持稳定的低等错误率(EER)(0.2%~1.0%),在SSTC2024数据集6种VC模型上也保持较低的EER;此外,FGE-SSV在未见过的VC模型上也展现出良好的泛化性能,其EER值均优于对比模型。

, authors=张国富1, 2, 3, 田天才1, 苏兆品1, 3, *, 王垚飞1, 3, 段宇衡1, 周康健1, 罗永琪1, authorsList=张国富, 田天才, 苏兆品, 王垚飞, 段宇衡, 周康健, 罗永琪, authorCompany=null, correspAuthors=苏兆品, authorNote=

作者贡献声明

张国富、田天才、苏兆品负责方案整体设计并撰写论文;王垚飞提出指导意见并修改论文;段宇衡、周康健、罗永琪负责部分数据集的制作。

1979年生。博士,教授。主要研究方向为人工智能、语音安全

2001年生。硕士研究生。主要研究方向为深伪语音检测与溯源

1983年生。博士,副教授。主要研究方向为网络与信息安全

1996年生。博士,副教授。主要研究方向为信息安全、人工智能安全

2004年生。本科生。主要研究方向为网络与信息安全

2005年生。本科生。主要研究方向为网络与信息安全

2005年生。本科生。主要研究方向为网络与信息安全

, correspAuthorsNote=
苏兆品()
, copyrightStatement=版权所有 © 2026 计算机研究与发展编辑部, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=77xA73shVn2LXZyo5suDuw==, magXml=Anrs5kq782drlpdx/fbPAw==, pdfUrl=null, pdf=x2MeVUxz0Kt9xU6IqcAdsw==, pdfFileSize=2171984, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=+ItVzjammg9E/W6eA+SsAQ==, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=wxEWJEixKKxQTaQ290DEfw==, mapNumber=null, fund=null)}, authors=[Author(id=1329420250799517865, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=zgf@hfut.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1329420250887598253, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, authorId=1329420250799517865, language=EN, stringName=Guofu Zhang, firstName=Guofu, middleName=null, lastName=Zhang, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, 3, address=1School of Computer Science and Information Engineering, Hefei University of Technology, Hefei 230601
2Institute of Interdisciplinary Science, Hefei University of Technology, Hefei 230009
3Intelligent Interconnected Systems Laboratory of Anhui Province (Hefei University of Technology), Hefei 230009, bio={"content":"

born in 1979. PhD, professor. His main research interests include artificial intelligence and audio security

"}, bioImg=null, bioContent=

born in 1979. PhD, professor. His main research interests include artificial intelligence and audio security

, aboutCorrespAuthor=null), CN=AuthorExt(id=1329420250963095726, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, authorId=1329420250799517865, language=CN, stringName=张国富, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, 3, address=1合肥工业大学计算机与信息学院 合肥 230601
2合肥工业大学交叉科学研究院 合肥 230009
3智能互联系统安徽省实验室(合肥工业大学) 合肥 230009, bio={"img":"oADyPtqGCemU6MKtRyyiDQ==","content":"

1979年生。博士,教授。主要研究方向为人工智能、语音安全

"}, bioImg=oADyPtqGCemU6MKtRyyiDQ==, bioContent=

1979年生。博士,教授。主要研究方向为人工智能、语音安全

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1329420250552053919, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, xref=1, ext=[AuthorCompanyExt(id=1329420250560442528, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, companyId=1329420250552053919, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1School of Computer Science and Information Engineering, Hefei University of Technology, Hefei 230601), AuthorCompanyExt(id=1329420250568831137, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, companyId=1329420250552053919, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1合肥工业大学计算机与信息学院 合肥 230601)]), AuthorCompany(id=1329420250635940002, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, xref=2, ext=[AuthorCompanyExt(id=1329420250640134307, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, companyId=1329420250635940002, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2Institute of Interdisciplinary Science, Hefei University of Technology, Hefei 230009), AuthorCompanyExt(id=1329420250648522916, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, companyId=1329420250635940002, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2合肥工业大学交叉科学研究院 合肥 230009)]), AuthorCompany(id=1329420250715631781, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, xref=3, ext=[AuthorCompanyExt(id=1329420250728214694, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, companyId=1329420250715631781, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3Intelligent Interconnected Systems Laboratory of Anhui Province (Hefei University of Technology), Hefei 230009), AuthorCompanyExt(id=1329420250736603303, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, companyId=1329420250715631781, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3智能互联系统安徽省实验室(合肥工业大学) 合肥 230009)])]), Author(id=1329420251042787504, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, orderNo=1, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1329420251126673586, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, authorId=1329420251042787504, language=EN, stringName=Tiancai Tian, firstName=Tiancai, middleName=null, lastName=Tian, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1School of Computer Science and Information Engineering, Hefei University of Technology, Hefei 230601, bio={"content":"

born in 2001. Master candidate. His main research interest includes deepfake audio detection and source tracing

"}, bioImg=null, bioContent=

born in 2001. Master candidate. His main research interest includes deepfake audio detection and source tracing

, aboutCorrespAuthor=null), CN=AuthorExt(id=1329420251202171059, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, authorId=1329420251042787504, language=CN, stringName=田天才, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1合肥工业大学计算机与信息学院 合肥 230601, bio={"img":"unNYqG/EeRiQ9jDWrLoOIA==","content":"

2001年生。硕士研究生。主要研究方向为深伪语音检测与溯源

"}, bioImg=unNYqG/EeRiQ9jDWrLoOIA==, bioContent=

2001年生。硕士研究生。主要研究方向为深伪语音检测与溯源

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1329420250552053919, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, xref=1, ext=[AuthorCompanyExt(id=1329420250560442528, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, companyId=1329420250552053919, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1School of Computer Science and Information Engineering, Hefei University of Technology, Hefei 230601), AuthorCompanyExt(id=1329420250568831137, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, companyId=1329420250552053919, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1合肥工业大学计算机与信息学院 合肥 230601)])]), Author(id=1329420251269279925, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, orderNo=2, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=szp@hfut.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=1, authorType=1, ext={EN=AuthorExt(id=1329420251348971704, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, authorId=1329420251269279925, language=EN, stringName=Zhaopin Su, firstName=Zhaopin, middleName=null, lastName=Su, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 3, *, address=1School of Computer Science and Information Engineering, Hefei University of Technology, Hefei 230601
3Intelligent Interconnected Systems Laboratory of Anhui Province (Hefei University of Technology), Hefei 230009, bio={"content":"

born in 1983. PhD, associate professor. Her main research interest includes network and information security

"}, bioImg=null, bioContent=

born in 1983. PhD, associate professor. Her main research interest includes network and information security

, aboutCorrespAuthor=null), CN=AuthorExt(id=1329420251411886265, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, authorId=1329420251269279925, language=CN, stringName=苏兆品, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 3, *, address=1合肥工业大学计算机与信息学院 合肥 230601
3智能互联系统安徽省实验室(合肥工业大学) 合肥 230009, bio={"img":"hqUf2xIpVH7zCp+eaEpW3w==","content":"

1983年生。博士,副教授。主要研究方向为网络与信息安全

"}, bioImg=hqUf2xIpVH7zCp+eaEpW3w==, bioContent=

1983年生。博士,副教授。主要研究方向为网络与信息安全

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1329420250552053919, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, xref=1, ext=[AuthorCompanyExt(id=1329420250560442528, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, companyId=1329420250552053919, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1School of Computer Science and Information Engineering, Hefei University of Technology, Hefei 230601), AuthorCompanyExt(id=1329420250568831137, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, companyId=1329420250552053919, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1合肥工业大学计算机与信息学院 合肥 230601)]), AuthorCompany(id=1329420250715631781, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, xref=3, ext=[AuthorCompanyExt(id=1329420250728214694, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, companyId=1329420250715631781, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3Intelligent Interconnected Systems Laboratory of Anhui Province (Hefei University of Technology), Hefei 230009), AuthorCompanyExt(id=1329420250736603303, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, companyId=1329420250715631781, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3智能互联系统安徽省实验室(合肥工业大学) 合肥 230009)])]), Author(id=1329420251483189435, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, orderNo=3, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1329420251554492606, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, authorId=1329420251483189435, language=EN, stringName=Yaofei Wang, firstName=Yaofei, middleName=null, lastName=Wang, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 3, address=1School of Computer Science and Information Engineering, Hefei University of Technology, Hefei 230601
3Intelligent Interconnected Systems Laboratory of Anhui Province (Hefei University of Technology), Hefei 230009, bio={"content":"

born in 1996. PhD, associate professor. His main research interests include information security and artificial intelligence security

"}, bioImg=null, bioContent=

born in 1996. PhD, associate professor. His main research interests include information security and artificial intelligence security

, aboutCorrespAuthor=null), CN=AuthorExt(id=1329420251617407167, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, authorId=1329420251483189435, language=CN, stringName=王垚飞, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 3, address=1合肥工业大学计算机与信息学院 合肥 230601
3智能互联系统安徽省实验室(合肥工业大学) 合肥 230009, bio={"img":"i3DCcX11nCbs3CnnvLsaKQ==","content":"

1996年生。博士,副教授。主要研究方向为信息安全、人工智能安全

"}, bioImg=i3DCcX11nCbs3CnnvLsaKQ==, bioContent=

1996年生。博士,副教授。主要研究方向为信息安全、人工智能安全

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1329420250552053919, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, xref=1, ext=[AuthorCompanyExt(id=1329420250560442528, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, companyId=1329420250552053919, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1School of Computer Science and Information Engineering, Hefei University of Technology, Hefei 230601), AuthorCompanyExt(id=1329420250568831137, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, companyId=1329420250552053919, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1合肥工业大学计算机与信息学院 合肥 230601)]), AuthorCompany(id=1329420250715631781, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, xref=3, ext=[AuthorCompanyExt(id=1329420250728214694, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, companyId=1329420250715631781, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3Intelligent Interconnected Systems Laboratory of Anhui Province (Hefei University of Technology), Hefei 230009), AuthorCompanyExt(id=1329420250736603303, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, companyId=1329420250715631781, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3智能互联系统安徽省实验室(合肥工业大学) 合肥 230009)])]), Author(id=1329420251692904641, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, orderNo=4, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1329420251785179333, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, authorId=1329420251692904641, language=EN, stringName=Yuheng Duan, firstName=Yuheng, middleName=null, lastName=Duan, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1School of Computer Science and Information Engineering, Hefei University of Technology, Hefei 230601, bio={"content":"

born in 2004. Undergraduate. His main research interest includes network and information security

"}, bioImg=null, bioContent=

born in 2004. Undergraduate. His main research interest includes network and information security

, aboutCorrespAuthor=null), CN=AuthorExt(id=1329420251885842630, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, authorId=1329420251692904641, language=CN, stringName=段宇衡, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1合肥工业大学计算机与信息学院 合肥 230601, bio={"img":"c3sNpAK7MKuoRrFUzlDBhw==","content":"

2004年生。本科生。主要研究方向为网络与信息安全

"}, bioImg=c3sNpAK7MKuoRrFUzlDBhw==, bioContent=

2004年生。本科生。主要研究方向为网络与信息安全

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1329420250552053919, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, xref=1, ext=[AuthorCompanyExt(id=1329420250560442528, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, companyId=1329420250552053919, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1School of Computer Science and Information Engineering, Hefei University of Technology, Hefei 230601), AuthorCompanyExt(id=1329420250568831137, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, companyId=1329420250552053919, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1合肥工业大学计算机与信息学院 合肥 230601)])]), Author(id=1329420251948757192, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, orderNo=5, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1329420252011671754, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, authorId=1329420251948757192, language=EN, stringName=Kangjian Zhou, firstName=Kangjian, middleName=null, lastName=Zhou, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1School of Computer Science and Information Engineering, Hefei University of Technology, Hefei 230601, bio={"content":"

born in 2005. Undergraduate. His main research interest includes network and information security

"}, bioImg=null, bioContent=

born in 2005. Undergraduate. His main research interest includes network and information security

, aboutCorrespAuthor=null), CN=AuthorExt(id=1329420252078780619, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, authorId=1329420251948757192, language=CN, stringName=周康健, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1合肥工业大学计算机与信息学院 合肥 230601, bio={"img":"Tn3YgdKBmC+LWMf0pjkJdg==","content":"

2005年生。本科生。主要研究方向为网络与信息安全

"}, bioImg=Tn3YgdKBmC+LWMf0pjkJdg==, bioContent=

2005年生。本科生。主要研究方向为网络与信息安全

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1329420250552053919, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, xref=1, ext=[AuthorCompanyExt(id=1329420250560442528, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, companyId=1329420250552053919, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1School of Computer Science and Information Engineering, Hefei University of Technology, Hefei 230601), AuthorCompanyExt(id=1329420250568831137, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, companyId=1329420250552053919, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1合肥工业大学计算机与信息学院 合肥 230601)])]), Author(id=1329420252158472397, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, orderNo=6, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1329420252229775567, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, authorId=1329420252158472397, language=EN, stringName=Yongqi Luo, firstName=Yongqi, middleName=null, lastName=Luo, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1School of Computer Science and Information Engineering, Hefei University of Technology, Hefei 230601, bio={"content":"

born in 2005. Undergraduate. His main research interest includes network and information security

"}, bioImg=null, bioContent=

born in 2005. Undergraduate. His main research interest includes network and information security

, aboutCorrespAuthor=null), CN=AuthorExt(id=1329420252296884432, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, authorId=1329420252158472397, language=CN, stringName=罗永琪, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1合肥工业大学计算机与信息学院 合肥 230601, bio={"img":"YkgW0ZPPFadRicAbzORurw==","content":"

2005年生。本科生。主要研究方向为网络与信息安全

"}, bioImg=YkgW0ZPPFadRicAbzORurw==, bioContent=

2005年生。本科生。主要研究方向为网络与信息安全

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1329420250552053919, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, xref=1, ext=[AuthorCompanyExt(id=1329420250560442528, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, companyId=1329420250552053919, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1School of Computer Science and Information Engineering, Hefei University of Technology, Hefei 230601), AuthorCompanyExt(id=1329420250568831137, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, companyId=1329420250552053919, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1合肥工业大学计算机与信息学院 合肥 230601)])])], keywords=[Keyword(id=1329420252393353425, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, language=EN, orderNo=1, keyword=source speaker verification), Keyword(id=1329420252456267986, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, language=EN, orderNo=2, keyword=voice conversion spoofing dataset), Keyword(id=1329420252519182547, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, language=EN, orderNo=3, keyword=feature generalised enhancement), Keyword(id=1329420252586291412, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, language=EN, orderNo=4, keyword=adversarial learning), Keyword(id=1329420252661788885, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, language=EN, orderNo=5, keyword=timbre feature), Keyword(id=1329420252728897750, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, language=CN, orderNo=1, keyword=源说话人验证), Keyword(id=1329420252816978135, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, language=CN, orderNo=2, keyword=语音转换伪造数据集), Keyword(id=1329420252879892696, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, language=CN, orderNo=3, keyword=特征泛化增强), Keyword(id=1329420252942807257, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, language=CN, orderNo=4, keyword=对抗学习), Keyword(id=1329420253009916122, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, language=CN, orderNo=5, keyword=音色特征)], refs=[Reference(id=1329420255941734655, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=1, rfOrder=0, authorNames=null, journalName=null, refType=null, unstructuredReference=李泽宇, 张旭鸿, 蒲誉文, 等. 多模态深度伪造及检测技术综述[J]. 计算机研究与发展, 2023, 60(6): 1396−1416, articleTitle=null, refAbstract=null), Reference(id=1329420256008843520, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=1, rfOrder=1, authorNames=null, journalName=null, refType=null, unstructuredReference=Li Zeyu, Zhang Xuhong, Pu Yuwen, et al. A survey on multimodal deepfake and detection techniques[J]. Journal of Computer Research and Development, 2023, 60(6): 1396−1416 (in Chinese), articleTitle=null, refAbstract=null), Reference(id=1329420256063369473, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=2, rfOrder=2, authorNames=null, journalName=null, refType=null, unstructuredReference=张晓辉, 易江燕, 陶建华, 等. 面向合成语音检测的弹性正交权重修正连续学习算法[J]. 计算机研究与发展, 2025, 62(2): 336−345, articleTitle=null, refAbstract=null), Reference(id=1329420256151449858, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=2, rfOrder=3, authorNames=null, journalName=null, refType=null, unstructuredReference=Zhang Xiaohui, Yi Jiangyan, Tao Jianhua, et al. Elastic orthogonal weight modification continual learning algorithm in the context of synthetic speech detection[J]. Journal of Computer Research and Development, 2025, 62(2): 336−345 (in Chinese), articleTitle=null, refAbstract=null), Reference(id=1329420256218558723, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=3, rfOrder=4, authorNames=null, journalName=null, refType=null, unstructuredReference=Liu Xuechen, Wang Xin, Sahidullah M, et al. ASVspoof 2021: Towards spoofed and deepfake speech detection in the wild[J]. IEEE/ACM Transactions on Audio, Speech, and Language Processing, 2023, 31: 2507−2522, articleTitle=null, refAbstract=null), Reference(id=1329420256285667588, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=4, rfOrder=5, authorNames=null, journalName=null, refType=null, unstructuredReference=Fan Cunhang, Ding Minging, Tao Jianhua, et al. Dual-branch knowledge distillation for noise-robust synthetic speech detection[J]. IEEE/ACM Transactions on Audio, Speech, and Language Processing, 2024, 32: 2453−2466, articleTitle=null, refAbstract=null), Reference(id=1329420256361165061, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=5, rfOrder=6, authorNames=null, journalName=null, refType=null, unstructuredReference=Wang Xin, Yamagishi J, Todisco M, et al. ASVspoof2019: A large-scale public database of synthesized, converted and replayed Speech[J]. Computer Speech & Language, 2020, 64: 101114, articleTitle=null, refAbstract=null), Reference(id=1329420256424079622, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=6, rfOrder=7, authorNames=null, journalName=null, refType=null, unstructuredReference=Yamagishi J, Wang Xin, Todisco M, et al. ASVspoof 2021: Accelerating progress in spoofed and deepfake speech detection[C]//Proc of the 4th Edition of the Automatic Speaker Verification and Spoofing Countermeasures Challenge. Red Hook, NY: Curran Associates Inc, 2021: 47−54, articleTitle=null, refAbstract=null), Reference(id=1329420256495382791, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=7, rfOrder=8, authorNames=null, journalName=null, refType=null, unstructuredReference=Pierno A D, Guarnera L, Allegra D, et al. End-to-end audio deepfake detection from RAW waveforms: A RawNet-based approach with cross-dataset evaluation[C]//Proc of Int Joint Conf on Neural Networks. Piscataway, NJ: IEEE, 2025: 1−8, articleTitle=null, refAbstract=null), Reference(id=1329420256554103048, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=8, rfOrder=9, authorNames=null, journalName=null, refType=null, unstructuredReference=McLaren M, Ferrer L, Castan D, et al. The speakers in the wild (SITW) speaker recognition database[C]//Proc of the 17th Annual Conf of the Int Speech Communication Association. Red Hook, NY: Curran Associates Inc, 2016: 818−822, articleTitle=null, refAbstract=null), Reference(id=1329420256604434697, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=9, rfOrder=10, authorNames=null, journalName=null, refType=null, unstructuredReference=Khalid H, Tariq S, Woo S S. FakeAVCeleb: A novel audio-video multimodal deepfake dataset[C]//Proc of the 35th Conf on Neural Information Processing Systems Track on Datasets and Benchmarks. Red Hook, NY: Curran Associates Inc, 2021: 1−14, articleTitle=null, refAbstract=null), Reference(id=1329420256667349258, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=10, rfOrder=11, authorNames=null, journalName=null, refType=null, unstructuredReference=Cai Danwei, Cai Zexin, Li Ming. Identifying source speakers for voice conversion based spoofing attacks on speaker verification systems[C]//Proc of the 48th IEEE Int Conf on Acoustics, Speech and Signal Processing. Piscataway, NJ: IEEE, 2023: 1−5, articleTitle=null, refAbstract=null), Reference(id=1329420256726069515, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=11, rfOrder=12, authorNames=null, journalName=null, refType=null, unstructuredReference=Deng Jiangyi, Chen Yanjiao, Zhong Yinan, et al. Catch you and I can: Revealing source voiceprint against voice conversion[C]//Proc of the 32nd USENIX Security Symp. Berkeley, CA: USENIX Association, 2023: 5163−5180, articleTitle=null, refAbstract=null), Reference(id=1329420256788984076, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=12, rfOrder=13, authorNames=null, journalName=null, refType=null, unstructuredReference=Ma Xinlei, Lu Wenhuan, Zhang Ruiteng, et al. Distillation-based feature extraction algorithm for source speaker verification[C]//Proc of IEEE Spoken Language Technology Workshop. Piscataway, NJ: IEEE, 2024: 1240−1246, articleTitle=null, refAbstract=null), Reference(id=1329420256872870157, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=12, rfOrder=14, authorNames=null, journalName=null, refType=null, unstructuredReference=Ma Xinlei, Lu Wenhuan, Zhang Ruiteng, et al. Distillation-based feature extraction algorithm for source speaker verification[C]//Proc of IEEE Spoken Language Technology Workshop. Piscataway, NJ: IEEE, 2024: 1240−1246, articleTitle=null, refAbstract=null), Reference(id=1329420256944173326, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=13, rfOrder=15, authorNames=null, journalName=null, refType=null, unstructuredReference=Li Ze, Lin Yuke, Yao Tian, et al. The database and benchmark for the source speaker tracing challenge 2024[C]//Proc of IEEE Spoken Language Technology Workshop. Piscataway, NJ: IEEE, 2024: 1254−1262, articleTitle=null, refAbstract=null), Reference(id=1329420257019670799, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=14, rfOrder=16, authorNames=null, journalName=null, refType=null, unstructuredReference=Panayotov V, Chen Guoguo, Povey D, et al. LibriSpeech: An ASR corpus based on public domain audio books[C]//Proc of the 19th IEEE Int Conf on Acoustics, Speech and Signal Processing. Piscataway, NJ: IEEE, 2015: 5206−5210, articleTitle=null, refAbstract=null), Reference(id=1329420257187442960, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=15, rfOrder=17, authorNames=null, journalName=null, refType=null, unstructuredReference=Nagrani A, Chung J S, Zisserman A. VoxCeleb: A large-scale speaker identification dataset[C]//Proc of the 18th Annual Conf of the Int Speech Communication Association. Red Hook, NY: Curran Associates Inc, 2017: 2616−2620, articleTitle=null, refAbstract=null), Reference(id=1329420257258746129, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=16, rfOrder=18, authorNames=null, journalName=null, refType=null, unstructuredReference=Liu Songting. Zero-shot voice conversion with diffusion transformers[J]. arXiv preprint, arXiv: 2411.09943, 2024, articleTitle=null, refAbstract=null), Reference(id=1329420257338437906, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=17, rfOrder=19, authorNames=null, journalName=null, refType=null, unstructuredReference=Ju Zeqian, Wang Yuancheng, Shen Kai, et al. NaturalSpeech 3: Zero-shot speech synthesis with factorized codec and diffusion models[C]//Proc of the 41st Int Conf on Machine Learning. Norfolk, MA: JMLR. org, 2024: 22605−22623, articleTitle=null, refAbstract=null), Reference(id=1329420257397158163, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=18, rfOrder=20, authorNames=null, journalName=null, refType=null, unstructuredReference=Choi H, Lee S, Lee S. DDDM-VC: Decoupled denoising diffusion models with disentangled representation and prior mixup for verified robust voice conversion[C]//Proc of the 38th AAAI Conf on Artificial Intelligence. Palo Alto, CA: AAAI, 2024: 17862−17870, articleTitle=null, refAbstract=null), Reference(id=1329420257460072725, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=19, rfOrder=21, authorNames=null, journalName=null, refType=null, unstructuredReference=Choi H, Lee S, Lee S. Diff-HierVC: Diffusion-based hierarchical voice Conversion with robust pitch generation and masked prior for zero-shot speaker adaptation[C]//Proc of the 24th Annual Conf of the Int Speech Communication Association. Red Hook, NY: Curran Associates Inc, 2023: 2283−2287, articleTitle=null, refAbstract=null), Reference(id=1329420257527181590, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=20, rfOrder=22, authorNames=null, journalName=null, refType=null, unstructuredReference=Park H J, Yang S W, Kim J S, et al. TriAAN-VC: Triple adaptive attention normalization for any-to-any voice conversion[C]//Proc of the 48th IEEE Int Conf on Acoustics, Speech and Signal Processing. Piscataway, NJ: IEEE, 2023: 1−5, articleTitle=null, refAbstract=null), Reference(id=1329420257602679063, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=21, rfOrder=23, authorNames=null, journalName=null, refType=null, unstructuredReference=Li Jingji, Tu Weiping, Xiao Li. FreeVC: Towards high-quality text-free one-shot voice conversion[C]//Proc of the 48th IEEE Int Conf on Acoustics, Speech and Signal Processing. Piscataway, NJ: IEEE, 2023: 1−5, articleTitle=null, refAbstract=null), Reference(id=1329420257669787928, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=22, rfOrder=24, authorNames=null, journalName=null, refType=null, unstructuredReference=Pan Xingang, Luo Ping, Shi Jianping, et al. Two at once: Enhancing learning and generalization capacities via IBN-Net[C]//Proc of the 15th European Conf on Computer Vision. Cham: Springer, 2018: 464−479, articleTitle=null, refAbstract=null), Reference(id=1329420257728508185, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=23, rfOrder=25, authorNames=null, journalName=null, refType=null, unstructuredReference=周俊佐, 易江燕, 陶建华, 等. 基于梅尔谱与压缩激励加权量化的语音神经编解码方法[J]. 计算机研究与发展, 2025, 62(11): 2725−2736, articleTitle=null, refAbstract=null), Reference(id=1329420257795617050, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=23, rfOrder=26, authorNames=null, journalName=null, refType=null, unstructuredReference=Zhou Junzuo, Yi Jiangyan, Tao Jianhua, et al. Neural speech codec method based on mel spectrogram and squeeze-excitation-weighted quantization[J]. Journal of Computer Research and Development, 2025, 62(11): 2725−2736 (in Chinese), articleTitle=null, refAbstract=null), Reference(id=1329420257858531611, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=24, rfOrder=27, authorNames=null, journalName=null, refType=null, unstructuredReference=苏兆品, 吴张倩, 岳峰, 等. 自然环境背景噪声下基于低维深度特征的手机来源识别[J]. 电子学报, 2021, 49(4): 637−646, articleTitle=null, refAbstract=null), Reference(id=1329420257917251868, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=24, rfOrder=28, authorNames=null, journalName=null, refType=null, unstructuredReference=Su Zhaopin, Wu Zhangqian, Yue feng, et al. Source cell-phone identification under background noise based on low-dimensional deep features[J] Acta Electronica Sinica, 2021, 49(4): 637−646 (in Chinese), articleTitle=null, refAbstract=null), Reference(id=1329420257996943645, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=25, rfOrder=29, authorNames=null, journalName=null, refType=null, unstructuredReference=Zhang Yin, Ye Mu, Zhu Guiyi, et al. FFCA-YOLO for small object detection in remote sensing images[J]. IEEE Transactions on Geoscience and Remote Sensing, 2024, 62: 5611215, articleTitle=null, refAbstract=null), Reference(id=1329420258059858206, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=26, rfOrder=30, authorNames=null, journalName=null, refType=null, unstructuredReference=Li Jiafeng, Wen Ying, He Lianghua. SCConv: Spatial and channel reconstruction convolution for feature redundancy[C]//Proc of IEEE/CVF Conf on Computer Vision and Pattern Recognition. Piscataway, NJ: IEEE, 2023: 6153−6162, articleTitle=null, refAbstract=null), Reference(id=1329420258135355679, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=27, rfOrder=31, authorNames=null, journalName=null, refType=null, unstructuredReference=Hu Jie, Shen Li, Sun Gang. Squeeze-and-excitation networks[C]//Proc of IEEE/CVF Conf on Computer Vision and Pattern Recognition. Piscataway, NJ: IEEE, 2018: 7132−7141, articleTitle=null, refAbstract=null), Reference(id=1329420258215047456, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=28, rfOrder=32, authorNames=null, journalName=null, refType=null, unstructuredReference=Chen Y, Wu D, Wu T, et al. Again-VC: A one-shot voice conversion using activation guidance and adaptive instance normalization[C]//Proc of the 46th IEEE Int Conf on Acoustics, Speech and Signal Processing. Piscataway, NJ: IEEE, 2021: 5954−5958, articleTitle=null, refAbstract=null), Reference(id=1329420258277962017, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=29, rfOrder=33, authorNames=null, journalName=null, refType=null, unstructuredReference=Gu Yewei, Zhang Zhenyu, Yi Xiaowei, et al. MediumVC: Any-to-any voice conversion using synthetic specific-speaker speeches as intermedium features[J]. arXiv preprint, arXiv: 2110.02500, 2021, articleTitle=null, refAbstract=null), Reference(id=1329420258353459490, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=30, rfOrder=34, authorNames=null, journalName=null, refType=null, unstructuredReference=Li Y A, Han Cong, Mesgarani N. StyleTTS: A style-based generative model for natural and diverse text-to-speech synthesis[J]. IEEE Journal of Selected Topics in Signal Processing, 2025, 19(1): 283−296, articleTitle=null, refAbstract=null), Reference(id=1329420258420568355, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=31, rfOrder=35, authorNames=null, journalName=null, refType=null, unstructuredReference=Wang Disong, Deng Liqun, Yeung Y T, et al. VQMIVC: Vector quantization and mutual information-based unsupervised speech representation disentanglement for one-shot voice conversion[C]//Proc of the 22nd Annual Conf of the Int Speech Communication Association. Red Hook, NY: Curran Associates Inc, 2021.1344−1348, articleTitle=null, refAbstract=null), Reference(id=1329420258496065828, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=32, rfOrder=36, authorNames=null, journalName=null, refType=null, unstructuredReference=Healy J, McInnes L. Uniform manifold approximation and projection[J]. Nature Reviews Methods Primers, 2024, 4(1): 82, articleTitle=null, refAbstract=null)], funds=null, companyList=[AuthorCompany(id=1329420250552053919, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, xref=1, ext=[AuthorCompanyExt(id=1329420250560442528, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, companyId=1329420250552053919, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1School of Computer Science and Information Engineering, Hefei University of Technology, Hefei 230601), AuthorCompanyExt(id=1329420250568831137, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, companyId=1329420250552053919, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1合肥工业大学计算机与信息学院 合肥 230601)]), AuthorCompany(id=1329420250635940002, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, xref=2, ext=[AuthorCompanyExt(id=1329420250640134307, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, companyId=1329420250635940002, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2Institute of Interdisciplinary Science, Hefei University of Technology, Hefei 230009), AuthorCompanyExt(id=1329420250648522916, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, companyId=1329420250635940002, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2合肥工业大学交叉科学研究院 合肥 230009)]), AuthorCompany(id=1329420250715631781, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, xref=3, ext=[AuthorCompanyExt(id=1329420250728214694, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, companyId=1329420250715631781, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3Intelligent Interconnected Systems Laboratory of Anhui Province (Hefei University of Technology), Hefei 230009), AuthorCompanyExt(id=1329420250736603303, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, companyId=1329420250715631781, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3智能互联系统安徽省实验室(合肥工业大学) 合肥 230009)])], figs=[ArticleFig(id=1329420253127356635, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, language=EN, label=Figure 1, caption=Illustration of sproofing speech detection and source speaker verification, figureFileSmall=ur/GfN0XRxsjhaF3e/ZPlA==, figureFileBig=+ItVzjammg9E/W6eA+SsAQ==, tableContent=null), ArticleFig(id=1329420253202854108, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, language=CN, label=图1, caption=伪造语音检测与源说话人验证示意图, figureFileSmall=ur/GfN0XRxsjhaF3e/ZPlA==, figureFileBig=+ItVzjammg9E/W6eA+SsAQ==, tableContent=null), ArticleFig(id=1329420253404180701, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, language=EN, label=Figure 2, caption=Illustration of data generation, figureFileSmall=gC60kMiavZWfE+XUK6Qw6Q==, figureFileBig=CkMeKu8koA/ukleBppRWKQ==, tableContent=null), ArticleFig(id=1329420253496455390, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, language=CN, label=图2, caption=数据生成示意图, figureFileSmall=gC60kMiavZWfE+XUK6Qw6Q==, figureFileBig=CkMeKu8koA/ukleBppRWKQ==, tableContent=null), ArticleFig(id=1329420253559369951, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, language=EN, label=Figure 3, caption=The overall framework of FGE-SSV model, figureFileSmall=myfSKCEs7Ii1o8qKbJT50Q==, figureFileBig=rqgBADYgxshE2cCseeM/jw==, tableContent=null), ArticleFig(id=1329420253613895904, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, language=CN, label=图3, caption=FGE-SSV模型总体框架, figureFileSmall=myfSKCEs7Ii1o8qKbJT50Q==, figureFileBig=rqgBADYgxshE2cCseeM/jw==, tableContent=null), ArticleFig(id=1329420253693587681, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, language=EN, label=Figure 4, caption=The FEM module structure, figureFileSmall=RQcrqWDL5Mw4S/7x5wYf6g==, figureFileBig=bfRxA6nAvNEExwYShtl+BQ==, tableContent=null), ArticleFig(id=1329420253752307938, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, language=CN, label=图4, caption=FEM模块结构, figureFileSmall=RQcrqWDL5Mw4S/7x5wYf6g==, figureFileBig=bfRxA6nAvNEExwYShtl+BQ==, tableContent=null), ArticleFig(id=1329420253811028195, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, language=EN, label=Figure 5, caption=Structure of IBN-SEBlock, figureFileSmall=EKZuO40IytRQ1aN/+j6kcA==, figureFileBig=XhK/rHskE2DCBYzSt7g72w==, tableContent=null), ArticleFig(id=1329420253878137060, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, language=CN, label=图5, caption=IBN- SEBlock结构, figureFileSmall=EKZuO40IytRQ1aN/+j6kcA==, figureFileBig=XhK/rHskE2DCBYzSt7g72w==, tableContent=null), ArticleFig(id=1329420253949440229, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, language=EN, label=Figure 6, caption=The overall structure of SCConv, figureFileSmall=2GS92C+39vkHQ2JfB2YjaQ==, figureFileBig=Q2kKVzmU5hgCBqFDFfxxbQ==, tableContent=null), ArticleFig(id=1329420254020743398, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, language=CN, label=图6, caption=SCConv总体结构, figureFileSmall=2GS92C+39vkHQ2JfB2YjaQ==, figureFileBig=Q2kKVzmU5hgCBqFDFfxxbQ==, tableContent=null), ArticleFig(id=1329420254083657959, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, language=EN, label=Figure 7, caption=The structure of SRU, figureFileSmall=MivEV7zMRDFUvrgQ4HeyAA==, figureFileBig=BuwNvz2nXJ6K+XFSJmMRFQ==, tableContent=null), ArticleFig(id=1329420254150766824, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, language=CN, label=图7, caption=SRU结构, figureFileSmall=MivEV7zMRDFUvrgQ4HeyAA==, figureFileBig=BuwNvz2nXJ6K+XFSJmMRFQ==, tableContent=null), ArticleFig(id=1329420254213681385, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, language=EN, label=Figure 8, caption=The structure of CRU, figureFileSmall=kUtNOzNllUYeHQ2syQAnLQ==, figureFileBig=6KgRsIhoQft/hDrVCu388g==, tableContent=null), ArticleFig(id=1329420254276595946, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, language=CN, label=图8, caption=CRU结构, figureFileSmall=kUtNOzNllUYeHQ2syQAnLQ==, figureFileBig=6KgRsIhoQft/hDrVCu388g==, tableContent=null), ArticleFig(id=1329420254339510507, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, language=EN, label=Figure 9, caption=The structure of Attention Merge, figureFileSmall=3BpUkOqbmfMMWz/Ao8BG6A==, figureFileBig=/HDbo/jBEy7BMSGbpmxmaQ==, tableContent=null), ArticleFig(id=1329420254406619372, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, language=CN, label=图9, caption=Attention Merge结构, figureFileSmall=3BpUkOqbmfMMWz/Ao8BG6A==, figureFileBig=/HDbo/jBEy7BMSGbpmxmaQ==, tableContent=null), ArticleFig(id=1329420254477922541, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, language=EN, label=Figure 10, caption=Visualization results of feature space distributions for FGE-SSV and VC-SID models across different datasets, figureFileSmall=nzAIvl8m+ug/PFpRgJddLQ==, figureFileBig=FD8IUeCpXThECJrbooJANg==, tableContent=null), ArticleFig(id=1329420254553420014, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, language=CN, label=图10, caption=FGE-SSV与VC-SID模型在各数据集上的特征空间分布可视化结果, figureFileSmall=nzAIvl8m+ug/PFpRgJddLQ==, figureFileBig=FD8IUeCpXThECJrbooJANg==, tableContent=null), ArticleFig(id=1329420254620528879, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, language=EN, label=Table 1, caption=

VCSD-SSV Dataset Structure

, figureFileSmall=null, figureFileBig=null, tableContent=
子集时长/
h
源说话
人数
(男/女)
目标说
话人数
(男/女)
源说话
人语
音数
VC语
音数
Seed-VC-LibriSpeech-
VoxCeleb1
145.6100/100100/1002308245558
FACodec-LibriSpeech-
VoxCeleb1
145.6100/100100/1002308245558
DDDM-VC-LibriSpeech-
VoxCeleb1
145.6100/100100/1002308245558
Diff-HierVC-LibriSpeech-
VoxCeleb1
145.6100/100100/1002308245558
TriAAN-VC-LibriSpeech-
VoxCeleb1
145.6100/100100/1002308245558
FreeVC-LibriSpeech-
VoxCeleb1
145.6100/100100/1002308245558
Source Speaker72.8100/10023082
Target Speaker85.2100/100
), ArticleFig(id=1329420254708609264, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, language=CN, label=表1, caption=

VCSD-SSV数据集结构

, figureFileSmall=null, figureFileBig=null, tableContent=
子集时长/
h
源说话
人数
(男/女)
目标说
话人数
(男/女)
源说话
人语
音数
VC语
音数
Seed-VC-LibriSpeech-
VoxCeleb1
145.6100/100100/1002308245558
FACodec-LibriSpeech-
VoxCeleb1
145.6100/100100/1002308245558
DDDM-VC-LibriSpeech-
VoxCeleb1
145.6100/100100/1002308245558
Diff-HierVC-LibriSpeech-
VoxCeleb1
145.6100/100100/1002308245558
TriAAN-VC-LibriSpeech-
VoxCeleb1
145.6100/100100/1002308245558
FreeVC-LibriSpeech-
VoxCeleb1
145.6100/100100/1002308245558
Source Speaker72.8100/10023082
Target Speaker85.2100/100
), ArticleFig(id=1329420254788301041, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, language=EN, label=Table 2, caption=

Experimental Configuration Environment

, figureFileSmall=null, figureFileBig=null, tableContent=
项目名称配置环境
CPUIntel Core i5-12600KF
RAM/GB8
GPUNVIDIA GeoForce RTX 4060Ti
CUDACuda 12.6
深度学习框架PyTorch 2.4
操作系统Windows 10
编程语言Python 3.9
), ArticleFig(id=1329420254863798514, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, language=CN, label=表2, caption=

实验配置环境

, figureFileSmall=null, figureFileBig=null, tableContent=
项目名称配置环境
CPUIntel Core i5-12600KF
RAM/GB8
GPUNVIDIA GeoForce RTX 4060Ti
CUDACuda 12.6
深度学习框架PyTorch 2.4
操作系统Windows 10
编程语言Python 3.9
), ArticleFig(id=1329420254935101683, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, language=EN, label=Table 3, caption=

Input-Output Dimension Configuration for Modules in FGE-SSV Model

, figureFileSmall=null, figureFileBig=null, tableContent=
模块输入维度输出维度
FEM(B,512,8,12)(B,512,8,12)
SCConv1(B,512,8,12)(B,512, 8,12)
IBN-SEBlock1(B,512,157)(B,512,157)
SCConv2(B,512,1,157)(B,512,157)
IBN-SEBlock2(B,512,157)(B,512,157)
SCConv3(B,512,1,157)(B,512,1,157)
IBN-SEBlock3(B,512,157)(B,512,157)
), ArticleFig(id=1329420254998016244, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, language=CN, label=表3, caption=

FGE-SSV模型模块的输入输出维度配置

, figureFileSmall=null, figureFileBig=null, tableContent=
模块输入维度输出维度
FEM(B,512,8,12)(B,512,8,12)
SCConv1(B,512,8,12)(B,512, 8,12)
IBN-SEBlock1(B,512,157)(B,512,157)
SCConv2(B,512,1,157)(B,512,157)
IBN-SEBlock2(B,512,157)(B,512,157)
SCConv3(B,512,1,157)(B,512,1,157)
IBN-SEBlock3(B,512,157)(B,512,157)
), ArticleFig(id=1329420255073513717, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, language=EN, label=Table 4, caption=

Ablation Experimental EER Results of FGE-SSV Modules on VCSD-SSV Dataset %

, figureFileSmall=null, figureFileBig=null, tableContent=
模型Seed-
VC
FACodecDDDM-
VC
Diff-HierVCTriAAN-
VC
FreeVC
FGE-SSV0.78.415.614.712.316.9
w/o FEM3.59.61715.612.918.3
w/o SCConv0.910.518.220.120.717.6
w/o IBN-SEBlock1.211.518.320.722.223
), ArticleFig(id=1329420255153205494, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, language=CN, label=表4, caption=

FGE-SSV模块在VCSD-SSV数据集上消融实验的EER结果

, figureFileSmall=null, figureFileBig=null, tableContent=
模型Seed-
VC
FACodecDDDM-
VC
Diff-HierVCTriAAN-
VC
FreeVC
FGE-SSV0.78.415.614.712.316.9
w/o FEM3.59.61715.612.918.3
w/o SCConv0.910.518.220.120.717.6
w/o IBN-SEBlock1.211.518.320.722.223
), ArticleFig(id=1329420255232897271, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, language=EN, label=Table 5, caption=

In-Domain Experimental Results for EER of Different Models on VCSD-SSV Dataset

, figureFileSmall=null, figureFileBig=null, tableContent=
VC模型FGE-SSV
(本文)
VC-SIDREVELIOSSTC
基准模型
Seed-VC0.7128.517.1
FACodec0.41.233.226.2
DDDM-VC0.22.128.414.3
Diff-HierVC0.71.89.911.3
TriAAN-VC13.443.819.9
FreeVC0.72.321.226.5
平均0.6227.519.2
), ArticleFig(id=1329420255320977656, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, language=CN, label=表5, caption=

各模型在VCSD-SSV数据集的域内EER实验结果 %

, figureFileSmall=null, figureFileBig=null, tableContent=
VC模型FGE-SSV
(本文)
VC-SIDREVELIOSSTC
基准模型
Seed-VC0.7128.517.1
FACodec0.41.233.226.2
DDDM-VC0.22.128.414.3
Diff-HierVC0.71.89.911.3
TriAAN-VC13.443.819.9
FreeVC0.72.321.226.5
平均0.6227.519.2
), ArticleFig(id=1329420255417446649, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, language=EN, label=Table 6, caption=

EER Results of Different Models on VCSD-SSV Dataset in Generalisation Experiment %

, figureFileSmall=null, figureFileBig=null, tableContent=
网络模型数据集Seed-VCFACodecDDDM-VCDiff-Hier VCTriAAN-VCFreeVC
VC-SIDSeed-VC-17.630.332.240.827.7
FACodec39.0-39.542.541.825.9
DDDM-VC30.822.0-4.736.433.8
Diff-HierVC31.628.82.5-52.236.1
TriAAN-VC37.632.739.241.5-27.8
FreeVC4345.248.247.145.3-
REVELIOSeed-VC-35.350.848.951.646.0
FACodec42.4-60.961.960.963.7
DDDM-VC35.031.0-31.331.432.5
Diff-HierVC19.715.713.6-21.324.4
TriAAN-VC52.151.951.851-51.6
FreeVC38.837.443.241.443.0-
SSTC
基准模型
Seed-VC-21.129.525.323.122.6
FACodec33.3-38.231.836.429.7
DDDM-VC27.920.2-17.829.826.9
Diff-HierVC28.722.915.0-21.133.9
TriAAN-VC30.227.725.520.1-31.6
FreeVC39.935.445.630.239.2-
FGE-SSV
(本文)
Seed-VC-8.415.614.712.316.9
FACodec26.1-32.734.527.926.6
DDDM-VC17.513.2-1.220.522.8
Diff-HierVC27.620.43.8-29.327.4
TriAAN-VC24.619.125.926.6-21.9
FreeVC33.726.739.539.230.0-
), ArticleFig(id=1329420255492944122, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, language=CN, label=表6, caption=

各模型在VCSD-SSV数据集上的泛化实验EER结果

, figureFileSmall=null, figureFileBig=null, tableContent=
网络模型数据集Seed-VCFACodecDDDM-VCDiff-Hier VCTriAAN-VCFreeVC
VC-SIDSeed-VC-17.630.332.240.827.7
FACodec39.0-39.542.541.825.9
DDDM-VC30.822.0-4.736.433.8
Diff-HierVC31.628.82.5-52.236.1
TriAAN-VC37.632.739.241.5-27.8
FreeVC4345.248.247.145.3-
REVELIOSeed-VC-35.350.848.951.646.0
FACodec42.4-60.961.960.963.7
DDDM-VC35.031.0-31.331.432.5
Diff-HierVC19.715.713.6-21.324.4
TriAAN-VC52.151.951.851-51.6
FreeVC38.837.443.241.443.0-
SSTC
基准模型
Seed-VC-21.129.525.323.122.6
FACodec33.3-38.231.836.429.7
DDDM-VC27.920.2-17.829.826.9
Diff-HierVC28.722.915.0-21.133.9
TriAAN-VC30.227.725.520.1-31.6
FreeVC39.935.445.630.239.2-
FGE-SSV
(本文)
Seed-VC-8.415.614.712.316.9
FACodec26.1-32.734.527.926.6
DDDM-VC17.513.2-1.220.522.8
Diff-HierVC27.620.43.8-29.327.4
TriAAN-VC24.619.125.926.6-21.9
FreeVC33.726.739.539.230.0-
), ArticleFig(id=1329420255564247291, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, language=EN, label=Table 7, caption=

EER Results of Different Models on SSTC Dataset in Generalisation Experiment %

, figureFileSmall=null, figureFileBig=null, tableContent=
网络模型数据集AGAIN-VCFreeVCMediumVCStyleTTSTriAAN-VCVQMIVC
VC-SIDAGAIN-VC24.5447.6647.6645.8448.1147.05
FreeVC48.1117.5237.8249.3245.6943.87
MediumVC46.4440.3925.8049.0246.2942.06
StyleTTS55.2247.3645.6621.346.7548.71
TriAAN-VC48.7141.9142.2148.1116.1642.97
VQMIVC56.5843.7239.7949.0244.9323.09
REVELIOAGAIN-VC40.348.350.848.951.656.0
FreeVC42.439.750.951.950.953.7
MediumVC48.449.338.648.245.547.9
StyleTTS49.247.741.640.542.854.4
TriAAN-VC47.553.452.351.540.653.2
VQMIVC55.158.555.452.851.743.2
SSTC
基准模型
AGAIN-VC22.6648.2647.9545.4746.4744.98
FreeVC41.617.5538.1639.0336.1636.16
MediumVC44.7840.3918.4642.0939.0345.04
StyleTTS45.0845.8447.1623.945.6950.08
TriAAN-VC42.9736.9140.2140.5417.743.12
VQMIVC47.544.4846.6144.6345.0833.13
FGE-SSV
(本文)
AGAIN-VC17.8244.3246.9243.1243.1746.6
FreeVC38.7114.2438.6137.234.7835.39
MediumVC46.1439.3215.7838.1135.8445.69
StyleTTS42.8439.4742.1120.738.2646.6
TriAAN-VC35.6940.5339.0935.8415.0343.27
VQMIVC46.639.9239.3237.6646.621.73
), ArticleFig(id=1329420255643939068, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, language=CN, label=表7, caption=

各模型在SSTC数据集上的泛化实验EER结果

, figureFileSmall=null, figureFileBig=null, tableContent=
网络模型数据集AGAIN-VCFreeVCMediumVCStyleTTSTriAAN-VCVQMIVC
VC-SIDAGAIN-VC24.5447.6647.6645.8448.1147.05
FreeVC48.1117.5237.8249.3245.6943.87
MediumVC46.4440.3925.8049.0246.2942.06
StyleTTS55.2247.3645.6621.346.7548.71
TriAAN-VC48.7141.9142.2148.1116.1642.97
VQMIVC56.5843.7239.7949.0244.9323.09
REVELIOAGAIN-VC40.348.350.848.951.656.0
FreeVC42.439.750.951.950.953.7
MediumVC48.449.338.648.245.547.9
StyleTTS49.247.741.640.542.854.4
TriAAN-VC47.553.452.351.540.653.2
VQMIVC55.158.555.452.851.743.2
SSTC
基准模型
AGAIN-VC22.6648.2647.9545.4746.4744.98
FreeVC41.617.5538.1639.0336.1636.16
MediumVC44.7840.3918.4642.0939.0345.04
StyleTTS45.0845.8447.1623.945.6950.08
TriAAN-VC42.9736.9140.2140.5417.743.12
VQMIVC47.544.4846.6144.6345.0833.13
FGE-SSV
(本文)
AGAIN-VC17.8244.3246.9243.1243.1746.6
FreeVC38.7114.2438.6137.234.7835.39
MediumVC46.1439.3215.7838.1135.8445.69
StyleTTS42.8439.4742.1120.738.2646.6
TriAAN-VC35.6940.5339.0935.8415.0343.27
VQMIVC46.639.9239.3237.6646.621.73
), ArticleFig(id=1329420255711047933, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, language=EN, label=Table 8, caption=

EER Results of Different Models on VCSD-SSV Dataset in Open-Set Generalisation Experiment %

, figureFileSmall=null, figureFileBig=null, tableContent=
网络模型数据集Seed-VCFACodecDDDM-VCDiff-Hier VCTriAAN-VCFreeVC
VC-SIDSeed-VC17.64550.750.249.240.4
FACodec54.920.352.551.148.639.4
DDDM-VC45.94417.443.739.746.7
Diff-HierVC51.243.945.316.161.647.2
TriAAN-VC50.041.146.145.017.837.2
FreeVC59.656.455.152.650.613.3
REVELIOSeed-VC34.861.563.250.358.553.5
FACodec56.140.766.56669.772.9
DDDM-VC53.253.338.654.248.847.0
Diff-HierVC37.235.844.713.352.042.8
TriAAN-VC55.056.257.655.346.863.0
FreeVC53.255.058.855.659.833.3
SSTC
基准模型
Seed-VC21.048.749.044.744.242.3
FACodec51.229.545.343.051.539.6
DDDM-VC46.538.820.149.849.633.8
Diff-HierVC47.040.541.114.557.341.6
TriAAN-VC52.041.854.248.229.339.7
FreeVC52.540.351.650.948.132.3
FGE-SSV
(本文)
Seed-VC14.939.847.946.336.640.8
FACodec42.617.544.242.348.535.6
DDDM-VC45.234.317.547.937.131.6
Diff-HierVC43.835.248.015.441.835.4
TriAAN-VC40.333.742.450.911.828.1
FreeVC48.138.954.642.241.811.5
), ArticleFig(id=1329420255803322622, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, language=CN, label=表8, caption=

各模型在VCSD-SSV数据集上的开集泛化实验EER结果

, figureFileSmall=null, figureFileBig=null, tableContent=
网络模型数据集Seed-VCFACodecDDDM-VCDiff-Hier VCTriAAN-VCFreeVC
VC-SIDSeed-VC17.64550.750.249.240.4
FACodec54.920.352.551.148.639.4
DDDM-VC45.94417.443.739.746.7
Diff-HierVC51.243.945.316.161.647.2
TriAAN-VC50.041.146.145.017.837.2
FreeVC59.656.455.152.650.613.3
REVELIOSeed-VC34.861.563.250.358.553.5
FACodec56.140.766.56669.772.9
DDDM-VC53.253.338.654.248.847.0
Diff-HierVC37.235.844.713.352.042.8
TriAAN-VC55.056.257.655.346.863.0
FreeVC53.255.058.855.659.833.3
SSTC
基准模型
Seed-VC21.048.749.044.744.242.3
FACodec51.229.545.343.051.539.6
DDDM-VC46.538.820.149.849.633.8
Diff-HierVC47.040.541.114.557.341.6
TriAAN-VC52.041.854.248.229.339.7
FreeVC52.540.351.650.948.132.3
FGE-SSV
(本文)
Seed-VC14.939.847.946.336.640.8
FACodec42.617.544.242.348.535.6
DDDM-VC45.234.317.547.937.131.6
Diff-HierVC43.835.248.015.441.835.4
TriAAN-VC40.333.742.450.911.828.1
FreeVC48.138.954.642.241.811.5
)], attaches=null, journal=Journal(id=1324759606947315908, delFlag=0, nameCn=计算机研究与发展, nameEn=Journal of Computer Research and Development, nameHistory1=null, nameHistory2=null, issn=1000-1239, eissn=null, cn=11-1777/TP, coden=null, periodic=0, language=CN, oaType=null, ccby=null, superviseOffice=null, ownerOffice=null, pubOffice=null, editorOffice=null, officeType=null, aims=null, clcCode=null, officeProv=null, officeCity=null, officeAddr=null, officeZip=null, officeEmail=null, officePhone=null, editDirector=null, officeDirector=null, officeDirectorPhone=null, officeStaffNum=null, officeEmpNum=null, coverPicUrl=k2l8zgqwstuxxevFX2qxVA==, journalPrice=null, startedYear=null, abbrevIsoEn=Journal of Computer Research and Development, journalRemark=null, publicationField=null, createdTime=1793776903141, updatedTime=1793850698612, createdBy=18614031015, updatedBy=13701087609, firstLetterCn=J, firstLetterEn=J, subjectCode=Electronic Information, subjectName=null, subjectCodeEn=Electronic Information, subjectNameEn=null, picCn=k2l8zgqwstuxxevFX2qxVA==, picEn=r6k0Ry+RD+c/VXj5QarmDw==, jcr=null, cjcr=null, exts=[JournalExt(id=1325069127710286688, language=CN, name=计算机研究与发展, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=, createdTime=1793850698640, updatedTime=1793850698640, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=, submissionAuthorUrl=https://crad.ict.ac.cn/journalx/authorLogOn.action, submissionEditorUrl=https://crad.ict.ac.cn/journalx/editorLogOn.action, submissionReviewUrl=https://crad.ict.ac.cn/journalx/expertLogOn.action, submissionCeEditorUrl=https://crad.ict.ac.cn/journalx/editorInChiefLogOn.action, submissionAeEditorUrl=, option={"copyright":""}), JournalExt(id=1325069127777395553, language=EN, name=Journal of Computer Research and Development, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=, createdTime=1793850698656, updatedTime=1793850698656, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=, submissionAuthorUrl=https://crad.ict.ac.cn/journalx/authorLogOn.action, submissionEditorUrl=https://crad.ict.ac.cn/journalx/editorLogOn.action, submissionReviewUrl=https://crad.ict.ac.cn/journalx/expertLogOn.action, submissionCeEditorUrl=https://crad.ict.ac.cn/journalx/editorInChiefLogOn.action, submissionAeEditorUrl=, option={"copyright":""})], databaseList=null, tenantJournalId=1324759914146529478, websiteList=[Website(id=1325070300139892771, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1324759914146529478, journalNameCn=null, journalNameEn=null, grayFlag=null, tenantId=1146029695717560320, platformId=null, journalGroupId=null, journalGroupNameCn=null, journalGroupNameEn=null, type=1, domain=https://castjournals.cast.org.cn/joweb/jsjyjyfz/CN, language=CN, createTime=1793850978169, createBy=18614031015, updateTime=1793851040259, updateBy=18614031015, name=计算机研究与发展-中文, tplId=1146099689490845704, title=计算机研究与发展, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1325070893751349360, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300139892771, code=articleTextType, value=kx, createTime=1793851119697, updateTime=1793851119697, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070893730377837, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300139892771, code=banner, value=null, createTime=1793851119692, updateTime=1793851119692, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070893772320883, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300139892771, code=grayFlag, value=0, createTime=1793851119702, updateTime=1793851119702, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070893721989228, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300139892771, code=logo, value=https://castjournals.cast.org.cn/joweb/jsjyjyfz/CN/file/pic?fileId=U27tDsy717YF7XSPA+Vwdw==, createTime=1793851119690, updateTime=1793851119690, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070893784903797, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300139892771, code=minRunFlag, value=0, createTime=1793851119705, updateTime=1793851119705, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070893742960751, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300139892771, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/jsjyjyfz/CN/file/pic, createTime=1793851119695, updateTime=1793851119695, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070893780709492, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300139892771, code=silenceFlag, value=0, createTime=1793851119704, updateTime=1793851119704, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070893738766446, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300139892771, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1793851119694, updateTime=1793851119694, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070893755543665, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300139892771, code=themeColor, value=null, createTime=1793851119698, updateTime=1793851119698, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070893763932274, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300139892771, code=themeStyle, value=null, createTime=1793851119700, updateTime=1793851119700, creator=18614031015, updator=18614031015)]), Website(id=1325070300580294698, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1324759914146529478, journalNameCn=null, journalNameEn=null, grayFlag=null, tenantId=1146029695717560320, platformId=null, journalGroupId=null, journalGroupNameCn=null, journalGroupNameEn=null, type=1, domain=https://castjournals.cast.org.cn/joweb/jsjyjyfz/EN, language=EN, createTime=1793850978274, createBy=18614031015, updateTime=1793851036638, updateBy=18614031015, name=计算机研究与发展-英文, tplId=1146101810881728533, title=Journal of Computer Research and Development, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1325070925288321148, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300580294698, code=articleTextType, value=kx, createTime=1793851127216, updateTime=1793851127216, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070925263155321, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300580294698, code=banner, value=null, createTime=1793851127210, updateTime=1793851127210, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070925309292671, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300580294698, code=grayFlag, value=0, createTime=1793851127221, updateTime=1793851127221, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070925254766712, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300580294698, code=logo, value=https://castjournals.cast.org.cn/joweb/jsjyjyfz/EN/file/pic?fileId=U27tDsy717YF7XSPA+Vwdw==, createTime=1793851127208, updateTime=1793851127208, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070925326069889, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300580294698, code=minRunFlag, value=0, createTime=1793851127225, updateTime=1793851127225, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070925279932539, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300580294698, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/jsjyjyfz/EN/file/pic, createTime=1793851127214, updateTime=1793851127214, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070925317681280, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300580294698, code=silenceFlag, value=0, createTime=1793851127223, updateTime=1793851127223, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070925271543930, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300580294698, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_en_623/, createTime=1793851127212, updateTime=1793851127212, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070925292515453, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300580294698, code=themeColor, value=null, createTime=1793851127217, updateTime=1793851127217, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070925300904062, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300580294698, code=themeStyle, value=null, createTime=1793851127219, updateTime=1793851127219, creator=18614031015, updator=18614031015)])], journalTitle=计算机研究与发展, weixinUrl=null, journalUrl=https://crad.ict.ac.cn/, iacademicId=null, status=1, seqNo=null, journalTitleEn=Journal of Computer Research and Development, journalPhotoCn=k2l8zgqwstuxxevFX2qxVA==, journalPhotoEn=r6k0Ry+RD+c/VXj5QarmDw==, journalFirstLetter=J, journalRecommend=null, journalNew=null, journalCollection=null, jcrJf=null, cjcrJf=null, jcrJfStr=null, cjcrJfStr=null, submissionFirstDecision=null, sciSubjectClassification=null, casSubjectClassification=null, citeScore=null, totalCitationFrequency=null, icpCode=null, psCode=null, advertisingLicenseCode=null, copyrightInformation=null, country=null, option=, provinceCode=null, provinceName=null, collectFlag=false, interPubPlatform=, interPubPlatformUrl=null), detailUrlCn=https://castjournals.cast.org.cn/joweb/jsjyjyfz/CN/10.7544/issn1000-1239.202550673, detailUrlEn=https://castjournals.cast.org.cn/joweb/jsjyjyfz/EN/10.7544/issn1000-1239.202550673, pdfUrlCn=https://castjournals.cast.org.cn/joweb/jsjyjyfz/CN/PDF/10.7544/issn1000-1239.202550673, pdfUrlEn=https://castjournals.cast.org.cn/joweb/jsjyjyfz/EN/PDF/10.7544/issn1000-1239.202550673, aliStartDate=0, aliEndDate=0, collectionFlag=false, citedCount=null, citedUrl=null, previewStatus=0, delFlag=0, hasFullText=1, orderTime=1785513600000, fullTextJson=null, articleText=null, reference=null)
收藏切换
面向源说话人验证的语音转换伪造数据集与模型
张国富 1, 2, 3 , 田天才 1 , 苏兆品 1, 3, * , 王垚飞 1, 3 , 段宇衡 1 , 周康健 1 , 罗永琪 1
计算机研究与发展 | 人工智能 2026,63(8): 2095-2110
收起
收藏切换
计算机研究与发展 |人工智能 2026 , 63 (8) : 2095 -2110
面向源说话人验证的语音转换伪造数据集与模型
全屏
[Author(id=1329420250799517865, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=zgf@hfut.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1329420250887598253, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, authorId=1329420250799517865, language=EN, stringName=Guofu Zhang, firstName=Guofu, middleName=null, lastName=Zhang, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, 3, address=1School of Computer Science and Information Engineering, Hefei University of Technology, Hefei 230601
2Institute of Interdisciplinary Science, Hefei University of Technology, Hefei 230009
3Intelligent Interconnected Systems Laboratory of Anhui Province (Hefei University of Technology), Hefei 230009, bio={"content":"

born in 1979. PhD, professor. His main research interests include artificial intelligence and audio security

"}, bioImg=null, bioContent=

born in 1979. PhD, professor. His main research interests include artificial intelligence and audio security

, aboutCorrespAuthor=null), CN=AuthorExt(id=1329420250963095726, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, authorId=1329420250799517865, language=CN, stringName=张国富, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, 3, address=1合肥工业大学计算机与信息学院 合肥 230601
2合肥工业大学交叉科学研究院 合肥 230009
3智能互联系统安徽省实验室(合肥工业大学) 合肥 230009, bio={"img":"oADyPtqGCemU6MKtRyyiDQ==","content":"

1979年生。博士,教授。主要研究方向为人工智能、语音安全

"}, bioImg=oADyPtqGCemU6MKtRyyiDQ==, bioContent=

1979年生。博士,教授。主要研究方向为人工智能、语音安全

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1329420250552053919, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, xref=1, ext=[AuthorCompanyExt(id=1329420250560442528, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, companyId=1329420250552053919, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1School of Computer Science and Information Engineering, Hefei University of Technology, Hefei 230601), AuthorCompanyExt(id=1329420250568831137, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, companyId=1329420250552053919, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1合肥工业大学计算机与信息学院 合肥 230601)]), AuthorCompany(id=1329420250635940002, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, xref=2, ext=[AuthorCompanyExt(id=1329420250640134307, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, companyId=1329420250635940002, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2Institute of Interdisciplinary Science, Hefei University of Technology, Hefei 230009), AuthorCompanyExt(id=1329420250648522916, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, companyId=1329420250635940002, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2合肥工业大学交叉科学研究院 合肥 230009)]), AuthorCompany(id=1329420250715631781, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, xref=3, ext=[AuthorCompanyExt(id=1329420250728214694, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, companyId=1329420250715631781, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3Intelligent Interconnected Systems Laboratory of Anhui Province (Hefei University of Technology), Hefei 230009), AuthorCompanyExt(id=1329420250736603303, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, companyId=1329420250715631781, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3智能互联系统安徽省实验室(合肥工业大学) 合肥 230009)])]), Author(id=1329420251042787504, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, orderNo=1, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1329420251126673586, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, authorId=1329420251042787504, language=EN, stringName=Tiancai Tian, firstName=Tiancai, middleName=null, lastName=Tian, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1School of Computer Science and Information Engineering, Hefei University of Technology, Hefei 230601, bio={"content":"

born in 2001. Master candidate. His main research interest includes deepfake audio detection and source tracing

"}, bioImg=null, bioContent=

born in 2001. Master candidate. His main research interest includes deepfake audio detection and source tracing

, aboutCorrespAuthor=null), CN=AuthorExt(id=1329420251202171059, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, authorId=1329420251042787504, language=CN, stringName=田天才, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1合肥工业大学计算机与信息学院 合肥 230601, bio={"img":"unNYqG/EeRiQ9jDWrLoOIA==","content":"

2001年生。硕士研究生。主要研究方向为深伪语音检测与溯源

"}, bioImg=unNYqG/EeRiQ9jDWrLoOIA==, bioContent=

2001年生。硕士研究生。主要研究方向为深伪语音检测与溯源

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1329420250552053919, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, xref=1, ext=[AuthorCompanyExt(id=1329420250560442528, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, companyId=1329420250552053919, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1School of Computer Science and Information Engineering, Hefei University of Technology, Hefei 230601), AuthorCompanyExt(id=1329420250568831137, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, companyId=1329420250552053919, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1合肥工业大学计算机与信息学院 合肥 230601)])]), Author(id=1329420251269279925, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, orderNo=2, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=szp@hfut.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=1, authorType=1, ext={EN=AuthorExt(id=1329420251348971704, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, authorId=1329420251269279925, language=EN, stringName=Zhaopin Su, firstName=Zhaopin, middleName=null, lastName=Su, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 3, *, address=1School of Computer Science and Information Engineering, Hefei University of Technology, Hefei 230601
3Intelligent Interconnected Systems Laboratory of Anhui Province (Hefei University of Technology), Hefei 230009, bio={"content":"

born in 1983. PhD, associate professor. Her main research interest includes network and information security

"}, bioImg=null, bioContent=

born in 1983. PhD, associate professor. Her main research interest includes network and information security

, aboutCorrespAuthor=null), CN=AuthorExt(id=1329420251411886265, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, authorId=1329420251269279925, language=CN, stringName=苏兆品, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 3, *, address=1合肥工业大学计算机与信息学院 合肥 230601
3智能互联系统安徽省实验室(合肥工业大学) 合肥 230009, bio={"img":"hqUf2xIpVH7zCp+eaEpW3w==","content":"

1983年生。博士,副教授。主要研究方向为网络与信息安全

"}, bioImg=hqUf2xIpVH7zCp+eaEpW3w==, bioContent=

1983年生。博士,副教授。主要研究方向为网络与信息安全

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1329420250552053919, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, xref=1, ext=[AuthorCompanyExt(id=1329420250560442528, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, companyId=1329420250552053919, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1School of Computer Science and Information Engineering, Hefei University of Technology, Hefei 230601), AuthorCompanyExt(id=1329420250568831137, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, companyId=1329420250552053919, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1合肥工业大学计算机与信息学院 合肥 230601)]), AuthorCompany(id=1329420250715631781, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, xref=3, ext=[AuthorCompanyExt(id=1329420250728214694, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, companyId=1329420250715631781, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3Intelligent Interconnected Systems Laboratory of Anhui Province (Hefei University of Technology), Hefei 230009), AuthorCompanyExt(id=1329420250736603303, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, companyId=1329420250715631781, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3智能互联系统安徽省实验室(合肥工业大学) 合肥 230009)])]), Author(id=1329420251483189435, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, orderNo=3, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1329420251554492606, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, authorId=1329420251483189435, language=EN, stringName=Yaofei Wang, firstName=Yaofei, middleName=null, lastName=Wang, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 3, address=1School of Computer Science and Information Engineering, Hefei University of Technology, Hefei 230601
3Intelligent Interconnected Systems Laboratory of Anhui Province (Hefei University of Technology), Hefei 230009, bio={"content":"

born in 1996. PhD, associate professor. His main research interests include information security and artificial intelligence security

"}, bioImg=null, bioContent=

born in 1996. PhD, associate professor. His main research interests include information security and artificial intelligence security

, aboutCorrespAuthor=null), CN=AuthorExt(id=1329420251617407167, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, authorId=1329420251483189435, language=CN, stringName=王垚飞, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 3, address=1合肥工业大学计算机与信息学院 合肥 230601
3智能互联系统安徽省实验室(合肥工业大学) 合肥 230009, bio={"img":"i3DCcX11nCbs3CnnvLsaKQ==","content":"

1996年生。博士,副教授。主要研究方向为信息安全、人工智能安全

"}, bioImg=i3DCcX11nCbs3CnnvLsaKQ==, bioContent=

1996年生。博士,副教授。主要研究方向为信息安全、人工智能安全

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1329420250552053919, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, xref=1, ext=[AuthorCompanyExt(id=1329420250560442528, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, companyId=1329420250552053919, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1School of Computer Science and Information Engineering, Hefei University of Technology, Hefei 230601), AuthorCompanyExt(id=1329420250568831137, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, companyId=1329420250552053919, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1合肥工业大学计算机与信息学院 合肥 230601)]), AuthorCompany(id=1329420250715631781, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, xref=3, ext=[AuthorCompanyExt(id=1329420250728214694, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, companyId=1329420250715631781, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3Intelligent Interconnected Systems Laboratory of Anhui Province (Hefei University of Technology), Hefei 230009), AuthorCompanyExt(id=1329420250736603303, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, companyId=1329420250715631781, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3智能互联系统安徽省实验室(合肥工业大学) 合肥 230009)])]), Author(id=1329420251692904641, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, orderNo=4, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1329420251785179333, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, authorId=1329420251692904641, language=EN, stringName=Yuheng Duan, firstName=Yuheng, middleName=null, lastName=Duan, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1School of Computer Science and Information Engineering, Hefei University of Technology, Hefei 230601, bio={"content":"

born in 2004. Undergraduate. His main research interest includes network and information security

"}, bioImg=null, bioContent=

born in 2004. Undergraduate. His main research interest includes network and information security

, aboutCorrespAuthor=null), CN=AuthorExt(id=1329420251885842630, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, authorId=1329420251692904641, language=CN, stringName=段宇衡, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1合肥工业大学计算机与信息学院 合肥 230601, bio={"img":"c3sNpAK7MKuoRrFUzlDBhw==","content":"

2004年生。本科生。主要研究方向为网络与信息安全

"}, bioImg=c3sNpAK7MKuoRrFUzlDBhw==, bioContent=

2004年生。本科生。主要研究方向为网络与信息安全

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1329420250552053919, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, xref=1, ext=[AuthorCompanyExt(id=1329420250560442528, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, companyId=1329420250552053919, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1School of Computer Science and Information Engineering, Hefei University of Technology, Hefei 230601), AuthorCompanyExt(id=1329420250568831137, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, companyId=1329420250552053919, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1合肥工业大学计算机与信息学院 合肥 230601)])]), Author(id=1329420251948757192, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, orderNo=5, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1329420252011671754, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, authorId=1329420251948757192, language=EN, stringName=Kangjian Zhou, firstName=Kangjian, middleName=null, lastName=Zhou, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1School of Computer Science and Information Engineering, Hefei University of Technology, Hefei 230601, bio={"content":"

born in 2005. Undergraduate. His main research interest includes network and information security

"}, bioImg=null, bioContent=

born in 2005. Undergraduate. His main research interest includes network and information security

, aboutCorrespAuthor=null), CN=AuthorExt(id=1329420252078780619, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, authorId=1329420251948757192, language=CN, stringName=周康健, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1合肥工业大学计算机与信息学院 合肥 230601, bio={"img":"Tn3YgdKBmC+LWMf0pjkJdg==","content":"

2005年生。本科生。主要研究方向为网络与信息安全

"}, bioImg=Tn3YgdKBmC+LWMf0pjkJdg==, bioContent=

2005年生。本科生。主要研究方向为网络与信息安全

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1329420250552053919, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, xref=1, ext=[AuthorCompanyExt(id=1329420250560442528, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, companyId=1329420250552053919, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1School of Computer Science and Information Engineering, Hefei University of Technology, Hefei 230601), AuthorCompanyExt(id=1329420250568831137, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, companyId=1329420250552053919, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1合肥工业大学计算机与信息学院 合肥 230601)])]), Author(id=1329420252158472397, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, orderNo=6, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1329420252229775567, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, authorId=1329420252158472397, language=EN, stringName=Yongqi Luo, firstName=Yongqi, middleName=null, lastName=Luo, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1School of Computer Science and Information Engineering, Hefei University of Technology, Hefei 230601, bio={"content":"

born in 2005. Undergraduate. His main research interest includes network and information security

"}, bioImg=null, bioContent=

born in 2005. Undergraduate. His main research interest includes network and information security

, aboutCorrespAuthor=null), CN=AuthorExt(id=1329420252296884432, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, authorId=1329420252158472397, language=CN, stringName=罗永琪, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1合肥工业大学计算机与信息学院 合肥 230601, bio={"img":"YkgW0ZPPFadRicAbzORurw==","content":"

2005年生。本科生。主要研究方向为网络与信息安全

"}, bioImg=YkgW0ZPPFadRicAbzORurw==, bioContent=

2005年生。本科生。主要研究方向为网络与信息安全

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1329420250552053919, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, xref=1, ext=[AuthorCompanyExt(id=1329420250560442528, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, companyId=1329420250552053919, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1School of Computer Science and Information Engineering, Hefei University of Technology, Hefei 230601), AuthorCompanyExt(id=1329420250568831137, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1329420246915592311, companyId=1329420250552053919, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1合肥工业大学计算机与信息学院 合肥 230601)])])]
张国富1, 2, 3 , 田天才1, 苏兆品1, 3, * , 王垚飞1, 3, 段宇衡1, 周康健1, 罗永琪1
作者信息
  • 1合肥工业大学计算机与信息学院 合肥 230601
  • 2合肥工业大学交叉科学研究院 合肥 230009
  • 3智能互联系统安徽省实验室(合肥工业大学) 合肥 230009
通讯作者:
苏兆品()
作者简介:

作者贡献声明

张国富、田天才、苏兆品负责方案整体设计并撰写论文;王垚飞提出指导意见并修改论文;段宇衡、周康健、罗永琪负责部分数据集的制作。

1979年生。博士,教授。主要研究方向为人工智能、语音安全

2001年生。硕士研究生。主要研究方向为深伪语音检测与溯源

1983年生。博士,副教授。主要研究方向为网络与信息安全

1996年生。博士,副教授。主要研究方向为信息安全、人工智能安全

2004年生。本科生。主要研究方向为网络与信息安全

2005年生。本科生。主要研究方向为网络与信息安全

2005年生。本科生。主要研究方向为网络与信息安全

Voice Conversion Spoofing Dataset and Model for Source Speaker Verification
Guofu Zhang1, 2, 3 , Tiancai Tian1, Zhaopin Su1, 3, * , Yaofei Wang1, 3, Yuheng Duan1, Kangjian Zhou1, Yongqi Luo1
Affiliations
  • 1School of Computer Science and Information Engineering, Hefei University of Technology, Hefei 230601
  • 2Institute of Interdisciplinary Science, Hefei University of Technology, Hefei 230009
  • 3Intelligent Interconnected Systems Laboratory of Anhui Province (Hefei University of Technology), Hefei 230009
  • born in 1979. PhD, professor. His main research interests include artificial intelligence and audio security

    born in 2001. Master candidate. His main research interest includes deepfake audio detection and source tracing

    born in 1983. PhD, associate professor. Her main research interest includes network and information security

    born in 1996. PhD, associate professor. His main research interests include information security and artificial intelligence security

    born in 2004. Undergraduate. His main research interest includes network and information security

    born in 2005. Undergraduate. His main research interest includes network and information security

    born in 2005. Undergraduate. His main research interest includes network and information security

出版时间: 2026-08-01 doi: 10.7544/issn1000-1239.202550673
文章导航
收藏切换

随着基于深度学习的语音转换(voice conversion,VC)技术的快速发展,攻击者能够通过篡改源说话人的语音信号来模仿目标说话人的声学特征,这严重威胁自动说话人验证(automatic speaker verification,ASV)系统的安全性。现有反欺骗防御机制主要聚焦于区分真实语音和伪造语音,却未能充分利用伪造语音中隐含的源说话人身份信息,导致其在司法取证、欺诈溯源等实际应用场景中难以满足需求。目前,针对源说话人身份验证与辨别的研究尚不充分,且已有模型泛化性能有限。基于此,构建了一个专门用于源说话人验证的新语音转换伪造数据集(voice conversion spoofing dataset for source speaker verification,VCSD-SSV)。该数据集涵盖Seed-VC,FACodec,DDDM-VC,Diff-HierVC,TriAAN-VC,Free-VC共6种语音转换算法,包含400个说话人,总计327 677条语音样本(其中真实语音46 811条、转换语音280 866条)。在此基础上,设计了一种基于特征泛化增强的源说话人验证(source speaker verification based on feature generalised enhancement,FGE-SSV)模型。该模型通过特征增强模块和级联的实例-批量归一化挤压激励块实现多层次源说话人音色特征的提取。实验结果表明,FGE-SSV在VCSD-SSV数据集全部6种VC模型中域内结果保持稳定的低等错误率(EER)(0.2%~1.0%),在SSTC2024数据集6种VC模型上也保持较低的EER;此外,FGE-SSV在未见过的VC模型上也展现出良好的泛化性能,其EER值均优于对比模型。

源说话人验证  /  语音转换伪造数据集  /  特征泛化增强  /  对抗学习  /  音色特征

With the rapid development of deep learning-based voice conversion (VC) technology, attackers can compromise the security of automatic speaker verification (ASV) systems by tampering with the source speaker’s speech signal to mimic the target speaker’s acoustic characteristics. The existing anti-spoofing countermeasures primarily focus on distinguishing between genuine and spoofed speech, yet underutilize the implicit source speaker identity information concealed within the spoofed speech. This limitation hinders their ability to meet practical demands in scenarios such as forensic investigation and fraud traceability. Currently, research on source speaker identity verification and discrimination remains insufficient, and existing methods suffer from limited generalisation capabilities. To address this, we construct a new voice conversion spoofing dataset specifically designed for source speaker verification, named VCSD-SSV (voice conversion spoofing dataset for source speaker verification). This dataset encompasses six VC algorithms (i.e., Seed-VC, FACodec, DDDM-VC, Diff-HierVC, TriAAN-VC, and Free-VC) and includes 400 speakers, totaling 327 677 speech samples (46 811 genuine samples and 280 866 converted samples). Building upon this, we propose a feature generalisation enhancement-based source speaker verification model, FGE-SSV (source speaker verification based on feature generalised enhancement). This model extracts multi-level source speaker timbre features through a feature enhancement module and cascaded instance-batch normalization squeeze-and-excitation blocks. Experimental results demonstrate that FGE-SSV maintains consistently low equal error rate (EER) (0.2%~1.0%) in within-domain results across all six VC methods in VCSD-SSV dataset and also achieves relatively low EER across the six VC methods in SSTC2024 dataset. Furthermore, FGE-SSV demonstrates strong generalization performance on unseen voice conversion methods, with its EER values outperforming those of comparative methods.

source speaker verification  /  voice conversion spoofing dataset  /  feature generalised enhancement  /  adversarial learning  /  timbre feature
张国富, 田天才, 苏兆品, 王垚飞, 段宇衡, 周康健, 罗永琪. 面向源说话人验证的语音转换伪造数据集与模型. 计算机研究与发展, 2026 , 63 (8) : 2095 -2110 . DOI: 10.7544/issn1000-1239.202550673
Guofu Zhang, Tiancai Tian, Zhaopin Su, Yaofei Wang, Yuheng Duan, Kangjian Zhou, Yongqi Luo. Voice Conversion Spoofing Dataset and Model for Source Speaker Verification[J]. Journal of Computer Research and Development, 2026 , 63 (8) : 2095 -2110 . DOI: 10.7544/issn1000-1239.202550673
在语音识别技术日益普及的背景下,自动说话人验证(automatic speaker verification,ASV)作为核心身份验证手段,已深度融入金融安全、智能设备认证等关键领域。然而,语音转换(voice conversion,VC)技术能够在保留原始语义内容的前提下,将源说话人的发音特性转换为目标说话人的音色特征。攻击者可借此篡改源说话人的声学特征,生成高度逼真的伪造语音,对ASV系统的安全性构成了严重威胁。
当前,多数反欺骗防御机制主要聚焦于伪造语音检测[1],即区分真实语音与伪造语音,却忽略了伪造语音中潜藏的源说话人身份信息。这导致其未能覆盖源说话人验证(source speaker verification,SSV)这一细粒度安全要求,在司法取证、欺诈溯源等实际应用场景中难以满足现实的迫切需求。
SSV不仅需提取语音中的常规声学特征,更需关注那些即使经过语音转换仍可能保留的源说话人特征,其与伪造语音检测的关系如图1所示。具体而言,伪造语音检测属于二分类任务,常利用深度学习算法区分真实语音和伪造语音[2-4]。该领域常用数据集包括:ASV2019[5],ASV2021[6],FoR(fake or real)[7],SITW(speakers in the wild)[8],FakeAVCeleb(fake audios and videos of celebrities)[9]等。
SSV的核心目标是从转换后的语音中准确提取源说话人信息。从技术层面看,现有语音转换技术普遍存在源说话人信息残留现象:由于声学特征解耦不完全或生成模型的固有约束,转换语音的频谱、韵律等特征中仍可能携带源说话人的生物标识,这为溯源攻击者身份提供了理论依据。
虽然已有模型[10-13]证明了在训练中引入标记源说话人身份的转换语音数据可在一定程度上实现源说话人识别,且利用多种不同VC模型生成的数据集训练有助于提升对未知转换模型的识别性能。但需要指出的是,不同VC模型在改变语音特征时对源说话人信息的保留和改变方式各有不同。面对性能不断提升的VC算法,如何高效准确地提取源说话人的音色特征信息,提升源说话人验证的泛化性,仍面临着严峻挑战。
基于上述背景,本文提出了面向源说话人验证的语音转换伪造数据集与模型。主要工作如下:
1)基于开源的LibriSpeech数据集[14]和VoxCeleb1数据集[15],采用最新发布的Seed-VC[16],FACodec[17],DDDM-VC[18],Diff-HierVC[19],TriAAN-VC[20],FreeVC[21] 等6种VC模型,构建了一个大规模源说话人验证语音转换伪造数据集(voice conversion spoofing dataset for source speaker verification,VCSD-SSV)。
2)提出了一种基于特征泛化增强的源说话人验证(source speaker verification based on feature generalised enhancement,FGE-SSV)模型。该模型利用特征增强模块和级联的实例-批量归一化挤压激励块(instance-batch normalization with sqeeze-and-excitation block,IBN-SEBlock)[22]实现多层次特征提取,并通过实验验证了所提模型的有效性。
本节从VC和SSV这2个方面对已有工作进行对比分析。
VC旨在改变语音的说话人身份(如共振峰、频谱包络、基频F0),同时保留语义内容,其架构包含训练与转换2阶段。训练阶段主要分析源/目标说话人语音,提取特征,训练映射模型建立两者关系;转换阶段聚焦提取源语音特征,利用映射模型生成目标特征,最终重构目标语音波形。现有VC技术主要分为以下3类:
1)编码-解码类。与传统语音编解码技术[23]不同的是,VC多采用Seq2Seq架构(编码器提取源特征,解码器生成目标语音),常需特征对齐。在这方面,FACodec[17]引入分解向量量化和分解扩散模型,实现人类自然度的语音生成。TriAAN-VC[20]使用3重自适应注意力归一化块,结合siamese损失,有效平衡源内容保持与目标说话人之间的相似性。FreeVC[21]基于VITS转换技术,利用信息瓶颈和数据增强从WavLM特征分离内容信息,实现高质量、鲁棒的一次性转换。
2)解耦类。侧重分离语音的语义内容与说话人身份,无需严格内容对齐。例如,DDDM-VC[18]将去噪过程分解为并行步骤来处理特定属性,提升可控性与质量。Diff-HierVC[19]分层系统分解为DiffPitch生成目标F0和DiffVoice转换语音。
3)扩散模型类。借鉴图像生成,通过噪声添加与去噪来学习复杂映射。例如,Seed-VC[16]基于零样本框架,引入音色变换器减少泄露,采用扩散变换器增强音色表征。Diff-HierVC[19]利用扩散模型生成F0并进行语音转换。
SSV是应对VC攻击下身份安全威胁的关键技术。与传统ASV主要依据声纹特征识别目标说话人不同,SSV旨在从经过深度转换的语音中鉴别出原始的源说话人身份,其核心科学问题是如何在目标说话人特征主导的信号中有效捕捉与分离微弱的源身份残留特征。当前该研究方向尚处于发展初期,仍面临公开数据集覆盖不足、VC模型泛化能力弱等挑战。
近年来,国内外学者已尝试从不同技术路径推进SSV的发展,主要包括以下3类:
1)基于编码器设计与监督训练。例如,VC-SID模型[10]采用ResNet编码结构,直接利用转换语音及其对应的源身份标签进行训练,从而学习源说话人特征表示。VC-SID模型在相同VC条件下表现良好,但跨模型泛化能力有限。
2)侧重于特征解耦与表示学习。例如,REVELIO框架[11]通过差分纠正算法抑制目标说话人特征,进而提取与源身份相关的正交表示。MeMo模型[12]利用自监督学习与多头多模式知识蒸馏框架实现高效特征提取。
3)引入注意力机制,用于增强对源信息的捕捉能力。例如,SSTC基准模型[13]利用注意力机制聚焦源特征并减弱目标说话人干扰。
虽然上述研究初步验证了SSV的可行性,但在特征表示的鲁棒性、跨VC泛化能力以及可解释性等方面仍存在明显不足,反映出SSV任务本身的艰巨性。与上述已有工作不同的是,本文构建了一个大规模源说话人验证语音转换伪造数据集VCSD-SSV,提出了一种基于特征泛化增强的源说话人验证模型FGE-SSV,旨在进一步提升源说话人验证的泛化性。
我们使用开源的LibriSpeech[14]和VoxCeleb1数据集[15]构建我们的VCSD-SSV数据集。LibriSpeech数据集采集自有声书网站,包含7个子集。VoxCeleb1数据集则包含1 251位名人的超过100 000个语音片段,并提供开发集和测试集。
本文选用LibriSpeech作为源语音数据集,VoxCeleb1作为目标语音数据集。为兼顾说话人的多样性与代表性,我们从LibriSpeech的train-clean-100子集中分别选取100名男性说话人(11 553条语音)和 100名女性说话人(11 529条语音),这些说话人具有不同的年龄、地域和语气特征,分别保存至LM(LibriSpeech_Male)和LF(LibriSpeech_Female)文件夹。同样,从VoxCeleb1开发集选取100名男性说话人(10 890条语音)和100名女性说话人(9 110条语音)作为目标说话人,分别保存至VM(VoxCeleb1_Male)和VF(VoxCeleb1_Female)文件夹。
我们使用Seed-VC,FACodec,DDDM-VC,Diff-HierVC,TriAAN-VC,FreeVC共6种VC模型生成转换数据。根据源说话人与目标说话人的性别组合,转换方式分为4类:男到男(male to male,M2M)、男到女(male to female,M2F)、女到女(female to female,F2F)和女到男(female to male,F2M)。每种VC模型的数据生成流程如图2所示。
数据生成的具体方法描述如下:
1)M2M。将LM文件夹中的说话人语音,使用VC模型依次向VM文件夹中的说话人转换。即,LM1向VM1转换,LM2向VM2转换,以此类推,共形成100对源-目标说话人组合。对于每一对组合,使用源说话人的所有语音作为源输入,并从目标说话人中随机选取一条语音作为参考语音(目标语音),应用指定的VC模型进行转换。
2)M2F,F2M,F2F。均遵循与M2M相同的生成方式,分别对应LM→VF,LF→VM,LF→VF的转换路径。
在数据生成阶段,针对每种VC模型(Seed-VC,FACodec,DDDM-VC,Diff-HierVC,TriAAN-VC,FreeVC),我们创建一个独立的文件夹用于存储其生成的转换语音。该文件夹命名为“VC名称-LibriSpeech-VoxCeleb1”。每个文件夹内进一步划分为4个子文件夹,分别存放不同性别转换类型的语音:M2M、M2F、F2M和F2F。所有语音文件均采用.wav格式。
在每个子文件夹中,每个源-目标说话人对的转换语音存储在以源说话人ID-目标说话人ID命名的目录下。例如,目录名103-id10270表示源说话人ID为103,目标说话人ID为id10270,即由说话人103向id10270转换的结果。
以使用FreeVC将LM中的男性说话人103向VM中的男性说话人id10270转换为例,生成的语音文件路径结构如下:
FreeVC-LibriSpeech-VoxCeleb1/M2M/103-id10270,
├── 103-1240-0001-id10270.wav,
├── 103-1240-0002-id10270.wav,
├── 103-1240-0003-id10270.wav,
其中103是源说话人ID,1240是章节编号,0001~0003是语音序列号,id10270是目标说话人ID。
基于上述规范构建的VCSD-SSV数据集结构如表1所示。该数据集包含6种语音转换模型(Seed-VC,FACodec等)生成的子集和原始语音库。所有VC子集均基于相同的源数据(100名源/目标说话人、23 082条源语音),生成等量的转换语音(各145.6 h、45 558条),且转换语音数为源语音的2倍;原始语音库则分为独立的源说话人子集(72.8 h、100人)和目标说话人子集(85.2 h、100人),前者用于对比实验或基线训练,后者用于说话人特征提取或验证任务,两者均未参与转换过程。
总之,我们的VCSD-SSV数据集1包含两大模块:VC转换语音库(6种模型×4.55万条语音)和原始语音库(2.3万条源语音+目标语音独立库)。这种统一数据基底+多模型对比+原始数据隔离的实验设计,为VC和SSV任务提供了多样化的数据支持。
本文提出的基于特征泛化增强的源说话人验证方法FGE-SSV,其总体框架如图3所示。FGE-SSV首先对输入语音(自然语音与转换语音)提取Fbank特征[24],并通过特征融合模块生成融合特征。该融合特征随后输入并行双分支进行处理。
1)特征增强分支。依次通过特征增强模块(feature enhancement module,FEM)[25]、空间与通道重建卷积(spatial and channel reconstruction convolution,SCConv)模块[26]提取源说话人信息。
2)泛化学习分支。通过级联的IBN-SEBlock[22]、SCConv[26]以及双分支注意力融合(attention merge)模块实现多层次特征提取。
最终,融合3个IBN-SEBlock的输出特征输入到源说话人验证网络进行身份验证。
下面我们将详细介绍一些关键模块的设计。
输入语音(包括自然语音与转换语音)首先经过Fbank特征[24]提取。随后,通过特征融合模块对Fbank特征进行处理,具体包含1维卷积(Conv1d)、ReLU激活函数以及批量归一化(batch normalization,BN)操作,最终生成高阶融合特征。
源说话人身份还原任务具有高度复杂性,模型常面临学习源说话人信息不足的挑战。为更有效地捕捉转换语音中的源说话人信息并获取更丰富的特征,FGE-SSV引入了FEM[25]。FEM模块旨在从2个维度对源说话人信息进行强化,其整体结构如图4所示。
FEM的核心设计采用多分支卷积架构,以挖掘多样化的说话人信息,从而提升特征的丰富程度。同时,为拓展感受野并捕获更丰富的局部上下文信息,模块中引入了空洞卷积。这种设计的优势在于,能够在保持模型轻量化的前提下,使生成的特征图蕴含更丰富的源说话人信息。
在图4中,FEM的计算过程由式(1)~(4)描述:
$ {\boldsymbol{X}}_{1}=f_{\text{conv}}^{{}^{3\times 3}}[f_{\text{conv}}^{{}^{1\times 1}}(\boldsymbol{F})] \;,$
$ {\boldsymbol{X}}_{2}=f_{\text{atconv}}^{{}^{3\times 3}}\{f_{\text{conv}}^{{}^{3\times 1}}\{f_{\text{conv}}^{{}^{1\times 3}}[f_{\text{conv}}^{{}^{1\times 1}}(\boldsymbol{F})]\}\} \;,$
$ {\boldsymbol{X}}_{3}=f_{\text{atconv}}^{{}^{3\times 3}}\{f_{\text{conv}}^{{}^{1\times 3}}\{f_{\text{conv}}^{{}^{3\times 1}}[f_{\text{conv}}^{{}^{1\times 1}}(\boldsymbol{F})]\}\} \;,$
$ {\boldsymbol{Y}}_{\text{FEM}}=Cat({\boldsymbol X}_{1},{\boldsymbol X}_{2},{\boldsymbol X}_{3})\oplus f_{\text{conv}}^{\text{1}\times \text{1}}(\boldsymbol{F}) \;,$
其中$ f_{\text{conv}}^{{}^{1\times 1}} $,$ f_{\text{conv}}^{{}^{1\times 3}} $,$ f_{\text{conv}}^{{}^{3\times 1}} $,$ f_{\text{conv}}^{{}^{3\times 3}} $分别代表使用1×1,1×3,3×1,3×3卷积核的标准卷积运算;$ f_{\text{atconv}}^{{}^{3\times 3}} $表示空洞率为5的空洞卷积运算;Cat(·)为特征图拼接操作;$ \oplus $表示特征图的逐元素相加;F为输入特征图;X1,X2,X3分别为前3个分支经标准卷积及空洞卷积处理后的输出特征图;YFEM是FEM模块的输出特征图。
为降低模型对特定训练数据集的依赖,提升其在未见数据上的泛化能力,FGE-SSV引入了以下关键组件:IBN-SEBlock[22],SCConv[26]及双分支Attention Merge模块。
在包含批 BN的CNN中,外观变化引起的特征差异主要集中于浅层,而深层特征虽富含说话人信息,浅层也具备内容区分能力。设计IBN-SEBlock模块的目的就是为了保护深层区分能力,避免在网络末端添加实例归一化(instance normalization,IN),同时保留浅层说话人信息,在残差路径中保留部分BN特征。
如图5所示,残差路径的首个卷积层后,对一半通道应用BN,另一半应用IN。其中,IN应置于残差路径而非恒等路径,以保持恒等路径的优化特性;通道分割策略(BN/IN各半)旨在平衡泛化能力与说话人信息保留。这是因为IN有助于学习外观不变特征以利用数据集内差异大的音频,同时对其适度引入能有效保留关键说话人信息。为最大化IN的泛化潜力,将其置于残差连接后,确保外观信息的同时在残差和恒等路径中保留。此外,IBN-SEBlock模块集成了SENet[27],通过对通道特征进行自适应加权,缓解模型对特定通道的过度依赖,降低过拟合风险,进一步增强泛化能力与特征表达能力。
SCConv模块旨在减少CNN特征中的空间和通道冗余,从而压缩模型以及降低复杂度与计算成本,同时提升特征表示能力。这有助于模型学习源说话人的本质特征,提高对未知语音转换模型的源说话人身份还原能力和泛化能力。如图6所示,SCConv模块主要包含空间重建单元(spatial reconstruction unit,SRU)和通道重建单元(channel reconstruction unit,CRU)两个核心单元。
1) SRU
如图7所示,利用组归一化(group normalization,GN)层中的缩放系数β量化特征信息量(信息丰富区β值大)。通过计算权重Wβ和门控阈值处理输出权重分量W1和W2,将输入特征X解耦为富含信息的特征$ \boldsymbol{X}_{1}^{\boldsymbol{W}} $和信息匮乏(冗余)的特征$ \boldsymbol{X}_{2}^{\boldsymbol{W}} $。
${\boldsymbol W_{\boldsymbol \beta }} = \boldsymbol \beta _i/\sum \limits_{j = 1}^C {\boldsymbol \beta _j},\;i = 1,2, … ,C,$
$ \boldsymbol{W}=Gate(S igmoid({\boldsymbol{W}}_{{{\emph{β}}}}(GN(\boldsymbol{X})))) \;,$
$ \left\{\begin{aligned} & \boldsymbol{X}_{1}^{\boldsymbol{W}}={\boldsymbol{W}}_{1}\otimes \boldsymbol{X}\;,\\& \boldsymbol{X}_{2}^{\boldsymbol{W}}={\boldsymbol{W}}_{2}\otimes \boldsymbol{X}\text{,}\end{aligned}\right. $
其中Gate为阈值函数,C为通道维度,$ \otimes $表示逐元素相乘。
随后采用交叉重构操作而非简单相加,有效融合不同信息密度的特征,增强信息交互,生成空间细化特征XW:
$ \left\{\begin{aligned} & \boldsymbol{X}_{11}^{\boldsymbol{W}}\oplus \boldsymbol{X}_{22}^{\boldsymbol{W}}={\boldsymbol{X}}^{{{\boldsymbol{W}}_{1}}}\;,\\&\boldsymbol{X}_{21}^{\boldsymbol{W}}\oplus \boldsymbol{X}_{12}^{\boldsymbol{W}}={\boldsymbol{X}}^{{{\boldsymbol{W}}_{2}}}\;,\\&{\boldsymbol{X}}^{{{\boldsymbol{W}}_{1}}}\cup {\boldsymbol{X}}^{{{\boldsymbol{W}}_{2}}}={\boldsymbol{X}}^{\boldsymbol{W}}\text{,}\end{aligned}\right. $
其中$ \oplus $表示逐元素相加,$ \cup $表示拼接。
2) CRU
如图8所示,对空间细化特征XW进行通道维度的冗余削减。采用拆分-变换-融合策略,利用轻量级卷积提取代表性特征XSCConv,并高效处理冗余特征以降低计算成本。SCConv模块通过有效消除空间和通道维度的冗余信息,显著抑制转换语音中目标说话人的影响,提升模型泛化能力。
Attention Merge模块负责融合来自不同分支(如FEM和泛化学习分支)的特征张量。其核心是使用注意力机制动态调整输入特征的权重,以更灵活高效地综合信息,如图9所示。
大致流程描述如下:
1)通过独立线性投影层将增强特征Xenh和泛化特征Xgen的每个通道在时间维度上映射为标量注意力分数Senh和Sgen:
$ {\boldsymbol{S}}_{\text{enh}}=\frac{L({\boldsymbol{X}}_{\text{enh}})}{\sqrt{t}} \;,$
$ {\boldsymbol{S}}_{\text{gen}}=\frac{L({\boldsymbol{X}}_{\text{gen}})}{\sqrt{t}} \;,$
其中L(·)是线性映射函数,t为特征的时间维度。
2)使用Softmax函数对Senh和Sgen进行归一化,得到各通道的注意力权重Wenh和Wgen反映其重要性。
3)将注意力权重与原始特征X相乘进行加权,并通过全局平均池化压缩时间维度,得到加权值Venh和Vgen。
4)根据融合权重Venh和Vgen对输入特征进行加权融合,并经过线性变换得到最终融合输出Xmerge:
$ {\boldsymbol{X}}_{\text{merge}}=L({\boldsymbol{V}}_{\text{enh}}\times {\boldsymbol{X}}_{\text{enh}}+{\boldsymbol{V}}_{\text{gen}}\times {\boldsymbol{X}}_{\text{gen}}) \;。 $
Attention Merge模块能自适应调整融合策略,考虑多尺度特征信息,提升模型从语音信号中提取源说话人信息的能力。此外,Attention Merge模块的使用增强了模型的鲁棒性,有助于防止过拟合。
FGE-SSV模型的泛化学习分支按以下流程处理特征:
1)特征融合模块的输出分别输入FEM模块和IBN-SEBlock模块。
2)IBN-SEBlock模块的输出输入到SCConv模块进行空间和通道冗余削减。
3)SCConv模块的输出与FEM模块的输出一同输入Attention Merge模块进行自适应加权融合。
4)融合后的特征可继续输入后续的SCConv模块和IBN-SEBlock模块进行进一步处理。
5)处理后的特征通过源说话人验证网络输出结果。
在图3中,源说话人验证网络对输入特征执行以下后处理流程,以基于提取的说话人信息进行身份判别:
1)为捕获不同层次的特征信息,将IBN-SEBlock模块中3个子模块的输出在通道维度上进行拼接。
2)利用1维卷积对拼接后的特征进行深度融合。
3)采用注意力统计池化(attentive statistics pooling,ASP)层处理融合特征的通道维度,生成更具判别性的统计特征表示。
4)通过全连接层(fully connected layer,FC)将聚合后的特征映射为一个192维度的说话人嵌入向量。
5)使用AAM-Softmax损失函数进行模型训练。该损失在标准Softmax中引入了角度裕度,以增大类间差异和增强分类判别能力,从而优化模型性能。
6)在类别判别层(type discrimination layer,TDL)中计算生成的嵌入向量与目标说话人注册向量之间的余弦距离,并依据该距离判断输入转换语音是否来源于目标说话人。
本节阐述用于评估我们提出的FGE-SSV模型性能的实验方案。
为验证FGE-SSV模型的有效性,我们选取了VC-SID[10],REVELIO[11],SSTC基准模型[13]3种模型进行对比实验分析。为了增强训练数据的多样性和对抗性,提高模型对合成语音,特别是多种转换技术所生成欺骗语料的判别能力,我们引入SSTC2024官网数据集[13]。该数据集通过16种不同的VC模型构建,在LibriSpeech和VoxCeleb2的基础上合成了大规模、多方法、多说话人配对的高仿真语音语料。SSTC基准模型[13]在训练阶段使用了超过32万条转换语音,开发和测试集也分别包含1.4万和1.3万条样本。除了上述VC模型,我们选取了SSTC2024官网数据集train1~train6共6种VC模型进行对比实验:TriAAN-VC[20],FreeVC[21],AGAIN-VC[28],MediumVC[29],StyleTTS[30],VQMIVC[31]。实验配置环境见表2。
FGE-SSV各模块的输入输出维度配置如表3所示。需特别说明的是,当模块间存在维度转换需求时,需要通过特征增维或降维操作确保数据维度的兼容性。FGE-SSV采用Adam优化器,初始学习率设为0.001,配合0.97的权重衰减系数,并采用分阶段学习率调整策略以提升训练稳定性。AAM-SoftMax损失函数中的超参数损失函数缩放因子scale=30,损失函数间隔参数margin=0.2。
为确保实验的可靠性,采用分层抽样方法对VCSD-SSV数据集和SSTC2024官网数据集进行实验数据划分:从每个VC数据集中随机选取20组样本进行训练,包含F2F,F2M,M2F,M2M这4种类型各5组。训练集包含源语音、目标语音和转换语音各约200条/人,涵盖20个说话人。
在测试阶段,所有样本均以语音对形式呈现,每对包含转换语音和源语音各1条。测试集保证各VC算法的数据量均衡,并采用二元标注机制:当源语音与转换语音的说话人ID一致时,标注为1;当说话人ID不一致时,标注为0。
最终在每个VC算法上构建的训练集规模约为5000条样本,测试集为2000对样本(1对样本对应1条源语音和1条转换语音)。
为明确本文所设计实验的适用场景,本文将实验划分为以下2类:
1)闭集设定。训练集与测试集共享相同的说话人集合,所有测试说话人均已在训练阶段出现。本文5.1~5.4节中所报告的域内及泛化实验均基于此设定,旨在验证模型对已知说话人身份的溯源性能。
2)开集设定。训练集与测试集的说话人ID互不重叠,模型需针对未见过的说话人,判断其是否是某段转换语音的源说话人(该部分实验结果将在本文5.5节进行介绍)。
采用等错误率(equal error rate,EER)作为源说话人验证的核心评价指标。EER是当误识率(false acceptation rate,FAR)与拒识率(false rejection rate,FRR)相等时的性能度量,能够综合反映检测模型的鲁棒性和平衡性。
$ FAR(\theta )=\frac{{N}_{\text{FA}}(\theta )}{{N}_{\text{sf}}} ,$
$ FRR(\theta )=\frac{{N}_{\text{FR}}(\theta )}{{N}_{\text{bf}}} , $
$ EER=FAR(\theta )=FRR(\theta ) , $
其中θ为判定阈值,Nsf和Nbf分别表示伪造语音和真实语音的样本总数,NFA(θ)为模型检测分数大于θ的伪造语音样本数,NFR(θ)为模型检测分数小于或等于θ的真实语音样本数。EER值越小,表明模型的检测性能越优。
在本节中,我们将从多个角度汇报并讨论实验结果。
为量化评估FGE-SSV模型中各关键模块对消除转换语音中目标音色干扰及提升模型泛化能力的作用,表4给出了FGE-SSV各模块在VCSD-SSV数据集上的消融实验的EER结果。
从表4可以看出,移除FEM模块后,模型在Seed-VC(域内)和其他数据集上的EER分别上升2.8个百分点和1.2~1.4个百分点。这表明FEM模块显著提升了模型对语音特征的表示能力,尤其在域内数据上表现更为突出(EER升幅最大达2.8个百分点)。这一结果验证了特征增强分支的双重作用:既防止模型过度偏向泛化性优化,又确保了域内性能的稳定性。其次,禁用SCConv模块导致域内EER小幅上升0.2个百分点,而在跨VC数据集上的EER升幅更为显著(2.1~8.4个百分点)。其中,在Diff-HierVC和TriAAN-VC数据集上的EER分别上升5.4个百分点和8.4个百分点,表明SCConv模块对复杂语音转换场景(如层级扩散模型和对抗生成模型)的泛化能力具有关键作用。此外,移除IBN-SEBlock模块后,模型在域内EER仅上升0.5个百分点,但在跨数据集上的性能退化尤为明显(EER升幅3.1~9.9个百分点)。例如,在TriAAN-VC和FreeVC数据集上,EER分别上升9.9个百分点和6.1个百分点,说明IBN-SEBlock模块通过特征归一化和通道注意力机制,有效增强了模型对异构语音转换方法的适应能力。
总之,FGE-SSV在所有数据集上均取得最低EER,而任何模块的移除均会导致性能退化,其中IBN-SEBlock模块的缺失对跨数据集影响最大。
为定量评估FGE-SSV模型从转换语音中提取源说话人音色特征的能力,我们基于VCSD-SSV数据集设计了特征空间分布对比实验。每种VC模型选取4组典型说话人对(M2M,F2M,M2F,F2F),每组说话人对包含10条源语音及其对应转换语音样本。图10展示了FGE-SSV与VC-SID在6种VC数据集上的特征空间分布UMAP[32]可视化结果,其中红色表示M2M转换,蓝色代表F2M转换,绿色表示M2F转换,紫色代表F2F转换。
由图10可以看出,FGE-SSV中同颜色点集呈现显著聚类特征,而VC-SID存在明显离散现象,尤其在Diff-HierVC数据集上出现多个离群点。其次,FGE-SSV成功分离4类转换场景,在性别转换场景(F2M/M2F)仍保持良好聚类,证明其能有效克服音高变异干扰,而VC-SID在跨性别转换中出现特征重叠,尤其是在TriAAN-VC数据集的M2F转换中出现特征漂移。
上述实验结果表明,FGE-SSV通过多尺度特征提取机制,有效分离了语音转换过程中注入的目标说话人特征与保留的源说话人特征。即使在性别转换场景下,FGE-SSV模型仍能捕捉源说话人的声道等生理特征,这解释了其在F2M/M2F任务中的优异表现。面对不同VC算法(特别是基于扩散模型的Diff-HierVC),FGE-SSV表现出更强的适应性,这与消融实验中IBN-SEBlock模块的作用相印证。
为验证FGE-SSV模型在域内场景下的有效性,我们选取训练集与测试集均来自同一VC模型生成的转换语音。表5给出了4种模型在VCSD-SSV数据集上的域内EER对比。由表5可以看出,FGE-SSV保持稳定的低EER(0.2%~1.0%),在全部6种VC模型上的EER平均下降70个百分点,在DDDM-VC上优势最显著。REVELIO表现波动剧烈(9.9%~43.8%),尤其在TriAAN-VC上异常退化(EER为 43.8%)。对于Diff-HierVC,因其转换痕迹明显,4种模型表现较优。可能的原因是,FGE-SSV中的FEM模块能够有效捕捉VC模型的特定伪影(如Diff-HierVC的频带失真),IBN-SEBlock模块抑制了生成式VC(Diff-HierVC/TriAAN-VC)的过度平滑效应和对抗性扰动。
为测试所提出FGE-SSV模型的泛化性,用其中一个VC模型的转换语音进行训练,而在另外5种VC模型上进行测试,实验结果如表6和表7所示。
由表6可以看出,与已有方法相比,所提FGE-SSV模型在未见过的语音转换方法上展现出良好的泛化性能。具体来说,采用Seed-VC模型进行训练时,在其他5种模型的EER平均为13.58%,均低于VC-SID的29.72%和REVELIO的46.52%以及SSTC基准模型的24.32%;采用FACodec模型进行训练时,其他5种模型的EER平均为29.56%,均低于VC-SID的37.74%和REVELIO的57.96%以及SSTC基准模型的33.38%;采用DDDM-VC模型进行训练时,其他5种模型的EER平均为15.04%,均低于VC-SID的24.66%和REVELIO的32.24%以及SSTC基准模型的24.52%;采用Diff-HierVC模型进行训练时,其他5种模型的EER平均为21.70%,低于VC-SID的30.16%以及SSTC基准模型的24.32%但略高于REVELIO的18.94%;采用TriAAN-VC模型进行训练时,其他5种模型的EER平均为23.62%, 均低于VC-SID的35.76%和REVELIO的51.68%以及SSTC基准模型的27.02%;采用FreeVC模型进行训练时,其他5种模型的EER平均为33.82%,均低于VC-SID的45.76%和REVELIO的40.76%以及SSTC基准模型的38.06%。
由表7可以看出,所提FGE-SSV模型在未见过的语音转换方法上也展现出良好的泛化性能。具体来说,采用AGAIN-VC模型进行训练时,其他5种模型的EER平均为44.83%,均低于VC-SID的47.26%和REVELIO的51.12%以及SSTC基准模型的46.62%;采用FreeVC模型进行训练时,在其他5种模型的EER平均为36.94%,均低于VC-SID的44.96%和REVELIO的49.96%以及SSTC基准模型的38.22%;采用MediumVC模型进行训练时,其他5种模型的EER平均为41.02%,均低于VC-SID的44.84%和REVELIO的47.86%以及SSTC基准模型的42.26%;采用StyleTTS模型进行训练时,其他5种模型的EER平均为41.85%,均低于VC-SID的48.74%和REVELIO的46.94%以及SSTC基准模型的47.14%;采用TriAAN-VC模型进行训练时,在其他5种模型的EER平均为38.88%,均低于VC-SID的44.78%和REVELIO的53.00%以及SSTC基准模型的40.75%;采用VQMIVC模型进行训练时,在其他5种模型的EER平均为42.02%,均低于VC-SID的46.81%和REVELIO的54.70%以及SSTC基准模型的45.66%。
需要指出的是,本文FGE-SSV模型在SSTC2024官网数据集上的实验结果普遍比在VCSD-SSV数据集上的性能下降,这可能是因为数据集的构造方式不同。VCSD-SSV数据集是成对式构造,即1个源说话人对应1个目标说话人,而SSTC2024官网数据集是针对每段目标语音, 随机选取3个源语音样本进行语音转换,模拟3个不同攻击者对同一目标语音发起的攻击。而实验中所选数据是根据源说话人来进行选择,SSTC2024官网数据集中所选实验数据中1个源说话人中的语音样本会对应多个目标说话人,因此1对1(1个源说话人对应1个目标说话人)的语音样本会大幅减少,而有效数据较少可能导致模型训练从转换语音中提取源说话人信息的效果降低。
为测试所提FGE-SSV模型在开集设定下的有效性,用开集设定下的测试集进行测试,实验结果如表8所示。
从表8可以看出,与表6相比,在开集条件下,所提FGE-SSV模型无论在域内还是在未见过的语音转换方法上,泛化性能都出现明显的下降,表明SSV方法在开集条件下仍具有很大的挑战性。但与已有方法相比,所提FGE-SSV模型在域内依然保持着较低的EER,且依然对未见语音转换方法保持最优平均泛化性能。具体来说,采用Seed-VC模型进行训练时,在其他5种模型的EER平均为42.28%,均低于VC-SID的47.1%和REVELIO的57.40%以及SSTC基准模型的45.78%;采用FACodec模型进行训练时,在其他5种模型的EER平均为42.64%,均低于VC-SID的49.3%和REVELIO的66.24%以及SSTC基准模型的46.12%;采用DDDM-VC模型进行训练时,在其他5种模型的EER平均为39.22%,均低于VC-SID的44%和REVELIO的51.3%以及SSTC基准模型的43.7%;采用Diff-HierVC模型进行训练时,在其他5种模型的EER平均为40.84%,低于VC-SID的49.84%和REVELIO的42.5%以及SSTC基准模型的45.5%;采用TriAAN-VC模型进行训练时,在其他5种模型的EER平均为39.08%,均低于VC-SID的43.88%和REVELIO的57.42%以及SSTC基准模型的47.18%; 采用FreeVC模型进行训练时,在其他5种模型的EER平均为45.12%,均低于VC-SID的54.86%和REVELIO的56.48%以及SSTC基准模型的48.68%。
值得注意的是,所提FGE-SSV模型在开集情况下的EER明显提高,可能的原因是:受限于单卡RTX 4060Ti(8 GB)显存,训练时仅使用20名说话人(约5 000条语音)进行训练,所见说话人特征空间远小于常规闭集方案。说话人先验分布的稀疏性导致模型在遇见开集全新说话人时,无法充分匹配残留源特征,从而EER会明显上升。
本文提出的FGE-SSV模型在VCSD-SSV与SSTC2024官网数据集上均取得显著性能优势,其在训练效果、泛化能力及与基线方法的比较中表现突出。这主要得益于以下设计:FEM模块增强了源说话人特征的表达能力,IBN-SE与SCConv模块提升了模型泛化性能,而双分支注意力融合机制则有效兼顾了对关键声纹信息的聚焦与泛化特征的融合,从而确保模型能够提取区分度更高的源说话人特征。
然而,FGE-SSV模型在部分特定场景下仍存在局限,主要表现在以下2方面:
1)低信噪比环境。由于所用训练数据VCSD-SSV基于高质量纯净语音构建,模型在噪声干扰较强的真实场景中表现可能下降,说明其对声学环境变化的鲁棒性仍有提升空间。
2)音色极度相似的说话人。当源说话人与目标说话人音色高度接近(如双胞胎、同性别同年龄段)时,模型判别性能显著降低。分析表明,此类转换语音中目标说话人特征占据主导,源说话人声学痕迹残留极少,导致FEM模块难以有效捕捉和分离出微弱的源特征信息。
这些情况反映出当前模型在特征解耦与弱痕迹提取方面仍存在挑战,尤其是在声纹信息被严重掩蔽或覆盖的场景下,模型架构对极微弱源特征的感知与增强机制尚需进一步优化。
本文围绕语音转换数据集构建与源说话人验证方法展开探索性工作,尝试将人工智能技术应用于源说话人的验证工作中。主要工作包括:采集并构建了语音转换数据集VCSD-SSV,并对其进行了细致的标注、转录和清洗等预处理工作;提出了基于特征泛化增强的源说话人验证方法FGE-SSV。实验结果表明,FGE-SSV模型在VCSD-SSV数据集和SSTC2024官网数据集上的表现相对优于现有模型,其源说话人验证效果在准确度和泛化性方面都取得了一定提升。具体而言,FGE-SSV方法在VCSD-SSV数据集全部6种VC模型中域内结果保持稳定的低EER(0.2~1.0个百分点),相较于对比方法EER平均下降70个百分点,展现出明显的优势;泛化性方面, FGE-SSV模型在未见过的语音转换方法上展现出良好的泛化性能,均优于对比方法。FGE-SSV方法在SSTC2024官网数据集6种VC模型上也保持较低的EER(15.03~21.73个百分点),相较于SSTC2024基准模型平均下降20个百分点;泛化性方面,也具有明显优势。这些初步尝试在源说话人验证的高准确度和泛化性方面取得了一定进展,为自动说话人验证系统的保护提供了新的技术思路。
但是,当前方法侧重于在可控环境下验证模型核心架构的有效性,仍有一些问题需要进一步验证和完善。首先,VCSD-SSV数据集的规模相对有限,在方言、低信噪比及跨语言场景下的验证不足,可能影响模型的训练效果和泛化能力。其次,现有数据集对最新发展的VC模型覆盖可能不够全面,因此模型在遇到从未见过的VC模型的泛化性仍有待改进。此外,目前对模型性能的评估方法还不够系统,缺乏更科学、更细致的评价体系。这些局限性都可能对研究结果的全面性产生一定影响。针对当前研究的不足,未来工作拟从以下2个方面进行改进:首先,计划将VCSD-SSV数据集规模扩展至10个VC模型以上,并开展方言、低信噪比及跨语言场景下的验证,以提升模型的泛化能力;其次,将探索泛化性更强的源说话人验证方法,进一步提高FGE-SSV遇到未知VC模型时的泛化性。我们希望通过这些改进,能够逐步提升FGE-SSV的性能,使其更好地保护自动说话人验证系统。

参考文献 引证文献
排序方式:
1
李泽宇, 张旭鸿, 蒲誉文, 等. 多模态深度伪造及检测技术综述[J]. 计算机研究与发展, 2023, 60(6): 1396−1416
Li Zeyu, Zhang Xuhong, Pu Yuwen, et al. A survey on multimodal deepfake and detection techniques[J]. Journal of Computer Research and Development, 2023, 60(6): 1396−1416 (in Chinese)
2
张晓辉, 易江燕, 陶建华, 等. 面向合成语音检测的弹性正交权重修正连续学习算法[J]. 计算机研究与发展, 2025, 62(2): 336−345
Zhang Xiaohui, Yi Jiangyan, Tao Jianhua, et al. Elastic orthogonal weight modification continual learning algorithm in the context of synthetic speech detection[J]. Journal of Computer Research and Development, 2025, 62(2): 336−345 (in Chinese)
3
Liu Xuechen, Wang Xin, Sahidullah M, et al. ASVspoof 2021: Towards spoofed and deepfake speech detection in the wild[J]. IEEE/ACM Transactions on Audio, Speech, and Language Processing, 2023, 31: 2507−2522
4
Fan Cunhang, Ding Minging, Tao Jianhua, et al. Dual-branch knowledge distillation for noise-robust synthetic speech detection[J]. IEEE/ACM Transactions on Audio, Speech, and Language Processing, 2024, 32: 2453−2466
5
Wang Xin, Yamagishi J, Todisco M, et al. ASVspoof2019: A large-scale public database of synthesized, converted and replayed Speech[J]. Computer Speech & Language, 2020, 64: 101114
6
Yamagishi J, Wang Xin, Todisco M, et al. ASVspoof 2021: Accelerating progress in spoofed and deepfake speech detection[C]//Proc of the 4th Edition of the Automatic Speaker Verification and Spoofing Countermeasures Challenge. Red Hook, NY: Curran Associates Inc, 2021: 47−54
7
Pierno A D, Guarnera L, Allegra D, et al. End-to-end audio deepfake detection from RAW waveforms: A RawNet-based approach with cross-dataset evaluation[C]//Proc of Int Joint Conf on Neural Networks. Piscataway, NJ: IEEE, 2025: 1−8
8
McLaren M, Ferrer L, Castan D, et al. The speakers in the wild (SITW) speaker recognition database[C]//Proc of the 17th Annual Conf of the Int Speech Communication Association. Red Hook, NY: Curran Associates Inc, 2016: 818−822
9
Khalid H, Tariq S, Woo S S. FakeAVCeleb: A novel audio-video multimodal deepfake dataset[C]//Proc of the 35th Conf on Neural Information Processing Systems Track on Datasets and Benchmarks. Red Hook, NY: Curran Associates Inc, 2021: 1−14
10
Cai Danwei, Cai Zexin, Li Ming. Identifying source speakers for voice conversion based spoofing attacks on speaker verification systems[C]//Proc of the 48th IEEE Int Conf on Acoustics, Speech and Signal Processing. Piscataway, NJ: IEEE, 2023: 1−5
11
Deng Jiangyi, Chen Yanjiao, Zhong Yinan, et al. Catch you and I can: Revealing source voiceprint against voice conversion[C]//Proc of the 32nd USENIX Security Symp. Berkeley, CA: USENIX Association, 2023: 5163−5180
12
Ma Xinlei, Lu Wenhuan, Zhang Ruiteng, et al. Distillation-based feature extraction algorithm for source speaker verification[C]//Proc of IEEE Spoken Language Technology Workshop. Piscataway, NJ: IEEE, 2024: 1240−1246
Ma Xinlei, Lu Wenhuan, Zhang Ruiteng, et al. Distillation-based feature extraction algorithm for source speaker verification[C]//Proc of IEEE Spoken Language Technology Workshop. Piscataway, NJ: IEEE, 2024: 1240−1246
13
Li Ze, Lin Yuke, Yao Tian, et al. The database and benchmark for the source speaker tracing challenge 2024[C]//Proc of IEEE Spoken Language Technology Workshop. Piscataway, NJ: IEEE, 2024: 1254−1262
14
Panayotov V, Chen Guoguo, Povey D, et al. LibriSpeech: An ASR corpus based on public domain audio books[C]//Proc of the 19th IEEE Int Conf on Acoustics, Speech and Signal Processing. Piscataway, NJ: IEEE, 2015: 5206−5210
15
Nagrani A, Chung J S, Zisserman A. VoxCeleb: A large-scale speaker identification dataset[C]//Proc of the 18th Annual Conf of the Int Speech Communication Association. Red Hook, NY: Curran Associates Inc, 2017: 2616−2620
16
Liu Songting. Zero-shot voice conversion with diffusion transformers[J]. arXiv preprint, arXiv: 2411.09943, 2024
17
Ju Zeqian, Wang Yuancheng, Shen Kai, et al. NaturalSpeech 3: Zero-shot speech synthesis with factorized codec and diffusion models[C]//Proc of the 41st Int Conf on Machine Learning. Norfolk, MA: JMLR. org, 2024: 22605−22623
18
Choi H, Lee S, Lee S. DDDM-VC: Decoupled denoising diffusion models with disentangled representation and prior mixup for verified robust voice conversion[C]//Proc of the 38th AAAI Conf on Artificial Intelligence. Palo Alto, CA: AAAI, 2024: 17862−17870
19
Choi H, Lee S, Lee S. Diff-HierVC: Diffusion-based hierarchical voice Conversion with robust pitch generation and masked prior for zero-shot speaker adaptation[C]//Proc of the 24th Annual Conf of the Int Speech Communication Association. Red Hook, NY: Curran Associates Inc, 2023: 2283−2287
20
Park H J, Yang S W, Kim J S, et al. TriAAN-VC: Triple adaptive attention normalization for any-to-any voice conversion[C]//Proc of the 48th IEEE Int Conf on Acoustics, Speech and Signal Processing. Piscataway, NJ: IEEE, 2023: 1−5
21
Li Jingji, Tu Weiping, Xiao Li. FreeVC: Towards high-quality text-free one-shot voice conversion[C]//Proc of the 48th IEEE Int Conf on Acoustics, Speech and Signal Processing. Piscataway, NJ: IEEE, 2023: 1−5
22
Pan Xingang, Luo Ping, Shi Jianping, et al. Two at once: Enhancing learning and generalization capacities via IBN-Net[C]//Proc of the 15th European Conf on Computer Vision. Cham: Springer, 2018: 464−479
23
周俊佐, 易江燕, 陶建华, 等. 基于梅尔谱与压缩激励加权量化的语音神经编解码方法[J]. 计算机研究与发展, 2025, 62(11): 2725−2736
Zhou Junzuo, Yi Jiangyan, Tao Jianhua, et al. Neural speech codec method based on mel spectrogram and squeeze-excitation-weighted quantization[J]. Journal of Computer Research and Development, 2025, 62(11): 2725−2736 (in Chinese)
24
苏兆品, 吴张倩, 岳峰, 等. 自然环境背景噪声下基于低维深度特征的手机来源识别[J]. 电子学报, 2021, 49(4): 637−646
Su Zhaopin, Wu Zhangqian, Yue feng, et al. Source cell-phone identification under background noise based on low-dimensional deep features[J] Acta Electronica Sinica, 2021, 49(4): 637−646 (in Chinese)
25
Zhang Yin, Ye Mu, Zhu Guiyi, et al. FFCA-YOLO for small object detection in remote sensing images[J]. IEEE Transactions on Geoscience and Remote Sensing, 2024, 62: 5611215
26
Li Jiafeng, Wen Ying, He Lianghua. SCConv: Spatial and channel reconstruction convolution for feature redundancy[C]//Proc of IEEE/CVF Conf on Computer Vision and Pattern Recognition. Piscataway, NJ: IEEE, 2023: 6153−6162
27
Hu Jie, Shen Li, Sun Gang. Squeeze-and-excitation networks[C]//Proc of IEEE/CVF Conf on Computer Vision and Pattern Recognition. Piscataway, NJ: IEEE, 2018: 7132−7141
28
Chen Y, Wu D, Wu T, et al. Again-VC: A one-shot voice conversion using activation guidance and adaptive instance normalization[C]//Proc of the 46th IEEE Int Conf on Acoustics, Speech and Signal Processing. Piscataway, NJ: IEEE, 2021: 5954−5958
29
Gu Yewei, Zhang Zhenyu, Yi Xiaowei, et al. MediumVC: Any-to-any voice conversion using synthetic specific-speaker speeches as intermedium features[J]. arXiv preprint, arXiv: 2110.02500, 2021
30
Li Y A, Han Cong, Mesgarani N. StyleTTS: A style-based generative model for natural and diverse text-to-speech synthesis[J]. IEEE Journal of Selected Topics in Signal Processing, 2025, 19(1): 283−296
31
Wang Disong, Deng Liqun, Yeung Y T, et al. VQMIVC: Vector quantization and mutual information-based unsupervised speech representation disentanglement for one-shot voice conversion[C]//Proc of the 22nd Annual Conf of the Int Speech Communication Association. Red Hook, NY: Curran Associates Inc, 2021.1344−1348
32
Healy J, McInnes L. Uniform manifold approximation and projection[J]. Nature Reviews Methods Primers, 2024, 4(1): 82
2026年第63卷第8期
75
36
0
引用本文
BibTeX
文章信息
doi: 10.7544/issn1000-1239.202550673
  • 接收时间:2025-09-10
  • 首发时间:2026-11-17
  • 出版时间:2026-08-01
补充材料
相关文章
文章信息
作者
出版历史
  • 收稿日期:2025-09-10
  • 修回日期:2026-03-12
  • 录用日期:2026-04-03
基金
作者信息
    1合肥工业大学计算机与信息学院 合肥 230601
    2合肥工业大学交叉科学研究院 合肥 230009
    3智能互联系统安徽省实验室(合肥工业大学) 合肥 230009

通讯作者:

苏兆品()
参考文献
分享链接
https://castjournals.cast.org.cn/joweb/jsjyjyfz/CN/10.7544/issn1000-1239.202550673
分享至
全文二维码

扫描看全文

引用本文
BibTeX
本文的引用情况
2种不同金属材料的力学参数
科
Family
属数
Number of
genus
种数
Number of
species
占总种数比例
Percentage of
total species (%)
属
Genus
种数
Number of
species
占总种数比例
Percentage of total
species (%)
鹅膏菌科Amanitaceae 2 11 5.26 鹅膏菌属 Amanita 10 4.78
小菇科 Mycenaceae 2 12 5.74 丝盖伞属 Inocybe 5 2.39
多孔菌科 Polyporaceae 8 14 6.70 蜡蘑属 Laccaria 5 2.39
红菇科 Russulaceae 3 23 11.00 小皮伞属 Marasmius 6 2.87
小菇属 Mycena 11 5.26
光柄菇属 Pluteus 5 2.39
红菇属 Russula 17 8.13
栓菌属 Trametes 5 2.39
关闭全屏