[논문]A Unified Deep Learning Framework for Short-Duration Speaker Verification in Adverse Environments

Jung, Youngmoon; Choi, Yeunju; Lim, Hyungjun; Kim, Hoirin

doi:10.1109/access.2020.3025941

[해외논문] A Unified Deep Learning Framework for Short-Duration Speaker Verification in Adverse Environments 원문보기

IEEE access : practical research, open solutions, v.8, 2020년, pp.175448 - 175466

Jung, Youngmoon (Korea Advanced Institute of Science and Technology, School of Electrical Engineering, Daejeon, South Korea) , Choi, Yeunju (Korea Advanced Institute of Science and Technology, School of Electrical Engineering, Daejeon, South Korea) , Lim, Hyungjun (Korea Advanced Institute of Science and Technology, School of Electrical Engineering, Daejeon, South Korea) , Kim, Hoirin (Korea Advanced Institute of Science and Technology, School of Electrical Engineering, Daejeon, South Korea)

Abstract ▼ AI-Helper

Speaker verification (SV) has recently attracted considerable research interest due to the growing popularity of virtual assistants. At the same time, there is an increasing requirement for an SV system: it should be robust to short speech segments, especially in noisy and reverberant environments. In this paper, we consider one more important requirement for practical applications: the system should be robust to an audio stream containing long non-speech segments, where a voice activity detection (VAD) is not applied. To meet these two requirements, we introduce feature pyramid module (FPM)-based multi-scale aggregation (MSA) and self-adaptive soft VAD (SAS-VAD). We present the FPM-based MSA to deal with short speech segments in noisy and reverberant environments. Also, we use the SAS-VAD to increase the robustness to long non-speech segments. To further improve the robustness to acoustic distortions (i.e., noise and reverberation), we apply a masking-based speech enhancement (SE) method. We combine SV, VAD, and SE models in a unified deep learning framework and jointly train the entire network in an end-to-end manner. To the best of our knowledge, this is the first work combining these three models in a deep learning framework. We conduct experiments on Korean indoor (KID) and VoxCeleb datasets, which are corrupted by noise and reverberation. The results show that the proposed method is effective for SV in the challenging conditions and performs better than the baseline ${i}$-vector and deep speaker embedding systems.

참고문헌 (76)

Proc IEEE Workshop Autom Speech Recog and Understanding The Kaldi speech recognition toolkit povey 2011
Jongseo Sohn, Nam Soo Kim, Wonyong Sung. A statistical model-based voice activity detection. IEEE signal processing letters, vol.6, no.1, 1-3.

상세보기
Varga, A., Steeneken, H.J.M.. Assessment for automatic speech recognition: II. NOISEX-92: A database and an experiment to study the effect of additive noise on speech recognition systems. Speech communication, vol.12, no.3, 247-251.

상세보기
IEEE Transactions on Audio Speech and Language Processing A tandem algorithm for pitch estimation and voiced speech segregation hu 2010 10.1109/TASL.2010.2041110 18 2067

상세보기
J Mach Learn Res Visualizing data using t-SNE van der maaten 2008 9 2579

상세보기
Proc Adv Neural Inf Process Syst Autodiff Workshop Automatic differentiation in pytorch paszke 2017
10.1109/CVPR.2019.00482
10.1109/ICASSP.2019.8682611
Wang, Feng, Cheng, Jian, Liu, Weiyang, Liu, Haijun. Additive Margin Softmax for Face Verification. IEEE signal processing letters, vol.25, no.7, 926-930.

상세보기
10.1109/ASRU46091.2019.9003935
10.21437/Interspeech.2019-2177
10.21437/Interspeech.2017-950
10.21437/Odyssey.2018-11
10.1109/CVPR.2017.713
10.1007/978-3-319-46478-7_31
Proc 33rd Int Conf Mach Learn Large-margin softmax loss for convolutional neural networks liu 2016 507
10.1109/ICASSP.2012.6288857
10.21437/Interspeech.2018-1151
Buda, Mateusz, Maki, Atsuto, Mazurowski, Maciej A.. A systematic study of the class imbalance problem in convolutional neural networks. Neural networks : the official journal of the International Neural Network Society, vol.106, 249-259.

상세보기
10.1109/ICCV.2017.324
Ghosh, Prasanta Kumar, Tsiartas, Andreas, Narayanan, Shrikanth. Robust Voice Activity Detection Using Long-Term Signal Variability. IEEE transactions on audio, speech, and language processing, vol.19, no.3, 600-613.

상세보기
10.21437/Odyssey.2020-66
arXiv 2003 12266 Dual attention in time and frequency domain for voice activity detection lee 2020
10.1109/ICASSP40776.2020.9053823
10.21437/Interspeech.2018-2461
10.1109/ICSDA.2017.8384419
arXiv 2005 03867 Multi-task network for noise-robust keyword spotting and speaker verification using CTC-based soft VAD and global query attention jung 2020
arXiv 1510 08484 [cs] MUSAN: A music, speech, and noise corpus snyder 2015
10.1109/ICASSP.2017.7953152
Aurora working group: DSR front end LVCSR evaluation AU/384/02 pearce 2002
Dehak, Najim, Kenny, Patrick J, Dehak, Réda, Dumouchel, Pierre, Ouellet, Pierre. Front-End Factor Analysis for Speaker Verification. IEEE transactions on audio, speech, and language processing, vol.19, no.4, 788-798.

상세보기
Hansen, John H. L., Hasan, Taufiq. Speaker Recognition by Machines and Humans: A tutorial review. IEEE signal processing magazine, vol.32, no.6, 74-99.

상세보기
Wang, Shuai, Huang, Zili, Qian, Yanmin, Yu, Kai. Discriminative Neural Embedding Learning for Short-Duration Text-Independent Speaker Verification. IEEE/ACM transactions on audio, speech, and language processing, vol.27, no.11, 1686-1696.

상세보기
arXiv 2004 03194 Improving multi-scale aggregation using feature pyramid module for robust speaker verification of variable-duration utterances jung 2020
10.1109/ASRU46091.2019.9004029
arXiv 2004 02863 Meta-learning for short utterance speaker recognition with imbalance length pairs min kye 2020
10.21437/Interspeech.2019-2240
10.21437/Interspeech.2019-1496
Al-Ali, Ahmed Kamil Hasan, Dean, David, Senadji, Bouchra, Chandran, Vinod, Naik, Ganesh R.. Enhanced Forensic Speaker Verification Using a Combination of DWT and MFCC Feature Warping in the Presence of Noise and Reverberation Conditions. IEEE access : practical research, open solutions, vol.5, 15400-15413.

상세보기
10.1109/ICASSP.2017.7953192
Proc Int Conf Med Image Comput -Assist Intervent U-net: Convolutional networks for biomedical image segmentation ronneberger 2015 234
10.1109/ICSDA.2017.8384446
Xiao-Lei Zhang, DeLiang Wang. Boosting Contextual Information for Deep Neural Network Based Voice Activity Detection. IEEE/ACM transactions on audio, speech, and language processing, vol.24, no.2, 252-264.

상세보기
Proc INTERSPEECH Comparison of forced-alignment speech recognition and humans for generating reference VAD kraljevski 2015 2937
10.21437/Interspeech.2016-268
10.1109/SLT.2018.8639586
10.21437/Interspeech.2018-1158
10.1109/ICASSP.2015.7178861
Proc Conf Neural Inf Process Syst Residual networks behave like ensembles of relatively shallow networks veit 2016 550
10.21236/ADA613971
10.1109/ICASSP.2014.6854363
10.21437/Interspeech.2019-2357
10.1109/ICASSP.2016.7472652
Zhang, Chunlei, Koishida, Kazuhito, Hansen, John H. L.. Text-Independent Speaker Verification Based on Triplet Convolutional Neural Network Embeddings. IEEE/ACM transactions on audio, speech, and language processing, vol.26, no.9, 1633-1644.

상세보기
10.1109/ICASSP.2018.8461375
Proc INTERSPEECH A time delay neural network architecture for efficient modeling of long temporal contexts peddinti 2015 3214
Proc Int Conf Learn Represent Very deep convolutional networks for large-scale image recognition simonyan 2015
10.1109/CVPR.2016.90
10.21437/Interspeech.2018-1769
10.21437/Interspeech.2018-1545
Proc Odyssey Bayesian speaker verification with heavy-tailed priors kenny 2010 14
Ioffe, S.. Probabilistic Linear Discriminant Analysis. Lecture notes in computer science, vol.3954, 531-542.

상세보기
10.1109/ICASSP.2013.6639151
Proc INTERSPEECH Analysis of i-vector length normalization in speaker recognition systems garcia-romero 2011 10.21437/Interspeech.2011-53 249
Proc INTERSPEECH I-vector based speaker recognition on short utterances kanagasundaram 2011 10.21437/Interspeech.2011-58 2341
Zhang, Xingyu, Zou, Xia, Sun, Meng, Zheng, Thomas Fang, Jia, Chong, Wang, Yimin. Noise Robust Speaker Recognition Based on Adaptive Frame Weighting in GMM for i-Vector Extraction. IEEE access : practical research, open solutions, vol.7, 27874-27882.

상세보기
10.21437/Interspeech.2019-2195
Proc Odyssey Deep neural networks for extracting Baum-Welch statistics for speaker recognition kenny 2014 10.21437/Odyssey.2014-44 293
He, K., Zhang, X., Ren, S., Sun, J.. Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition. Lecture notes in computer science, vol.8691, 346-361.

상세보기
10.21437/Interspeech.2018-993
10.21437/Interspeech.2019-1489
Campbell, W.M., Sturim, D.E., Reynolds, D.A.. Support vector machines using GMM supervectors for speaker verification. IEEE signal processing letters, vol.13, no.5, 308-311.

상세보기
10.1109/ICCV.2019.00346
10.21437/Interspeech.2017-1575
10.21437/Interspeech.2017-620
10.1109/SLT.2016.7846260

LOADING...

활용도 분석정보

상세보기

다운로드

내보내기

활용도 Top5 논문

해당 논문의 주제분야에서 활용도가 높은 상위 5개 콘텐츠를 보여줍니다.
더보기 버튼을 클릭하시면 더 많은 관련자료를 살펴볼 수 있습니다.

원문 보기

AccessON 원문보기

원문 URL 링크

DOI : 10.1109/ACCESS.2020.3025941 [무료]
IEEE : 저널 > 논문 [무료]
AccessON : 저널

*원문 PDF 파일 및 링크정보가 존재하지 않을 경우 KISTI DDS 시스템에서 제공하는 원문복사서비스를 사용할 수 있습니다.

오픈액세스(OA) 유형

GOLD

오픈액세스 학술지에 출판된 논문

저작권 관리 안내

내보내기 메뉴

내보내기 구분

파일저장
인쇄
메일전송

구성항목

기본정보
상세정보

관리번호, 논문명, 저널/프로시딩명, 저자 , 발행년, 권, 호, 시작페이지, 끝페이지, 발행기관

저장형식

Text(ASCII format)
Excel format
RefWorks Direct Export
RIS format (for Reference Manager, ProCite, EndNote), Scholar's Aids, Mendeley

메일정보

받는사람 (필수): @
보내는사람 (선택): @
제목
내용: KISTI 검색결과 이메일 서비스

안내

총 건의 자료가 검색되었습니다.

다운받으실 자료의 인덱스를 입력하세요. (1-10,000)

검색결과의 순서대로 최대 10,000건 까지 다운로드가 가능합니다.

데이타가 많을 경우 속도가 느려질 수 있습니다.(최대 2~3분 소요)

다운로드 파일은 UTF-8 형태로 저장됩니다.
파일의 내용이 제대로 보이지 않을실 때는 웹브라우저 상단의 보기 -> 인코딩 -> 자동선택 여부를 확인하십시오.

Text(ASCII format)
Excel format

표제어: PCR

동의어: Packet Collision Rate

용어 설명 출처 목록 (6)

용어 설명: PCR은 세균 특이성이 있는 primer를 이용하여 적은 수의 세균이 있을지라도 쉽게 검출할 수 있는 유용한 방법이며, 이를 이용하여 구강 내 치면세균막이나 타액에서 직접 세균을 검출할 수 있게 되었다[8].

AI-Helper ※ AI-Helper는 을 사용합니다.

AI-Helper

안녕하세요, AI-Helper입니다. 좌측 "선택된 텍스트"에서 텍스트를 선택하여 요약, 번역, 용어설명을 실행하세요.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.

연합인증