Speaker Tracking Using Eigendecomposition and an Index Tree of Reference Models

다국어 초록 (Multilingual Abstract)

This paper focuses on online speaker tracking for telephone conversations and broadcast news. Since the online applicability imposes some limitations on the tracking strategy, such as data insufficiency, a reliable approach should be applied to compensate for this shortage. In this framework, a set of reference speaker models are used as side information to facilitate online tracking. To improve the indexing accuracy, adaptation approaches in eigenvoice decomposition space are proposed in this paper. We believe that the eigenvoice adaptation techniques would help to embed the speaker space in the models and hence enrich the generality of the selected speaker models. Also, an index structure of the reference models is proposed to speed up the search in the model space. The proposed framework is evaluated on 2002 Rich Transcription Broadcast News and Conversational Telephone Speech corpus as well as a synthetic dataset. The indexing errors of the proposed framework on telephone conversations, broadcast news,and synthetic dataset are 8.77%, 9.36%, and 12.4%,respectively. Using the index tree structure approach, the run time of the proposed framework is improved by 22%.

참고문헌 (Reference)

1 S. Kwon, "Unsupervised Speaker Indexing Using Generic Models" 13 : 1004-1013, 2004

2 Y.K. Muthusamy, "The OGI Multi-language Telephone Speech Corpus" 895-898, 1992

3 C. Wooters, "The ICSI RT07s Speaker Diarization System" 2007

4 "The 2009 (RT-09) Rich Transcription Evaluation Plan"

5 M.Bijankhan,Great Farsdat Database, "Technical report" Iran Research Center on Intelligent Signal Processing 2002

6 M. Davy, "Supervised Classification Using MCMC Methods" 33-36, 2000

7 S. Meignier, "Step-by-Step and Integrated Approaches in Broadcast News Speaker Diarization" 20 (20): 303-330, 2006

8 D.A. Reynolds, "Speaker Verification Using Adapted Gaussian Mixture Models" 10 (10): 19-41, 2000

9 M. Kotti, "Speaker Segmentation and Clustering" 88 (88): 1091-1124, 2008

10 A. Martin, "Speaker Recognition in a Multispeaker Environment" 787-790, 2001

11 D.A. Reynolds, "Speaker Identification and Verification Using Gaussian Mixture Speaker Models" 17 (17): 91-108, 1995

12 K. Iso, "Speaker Clustering Using Vector Quantization and Spectral Clustering" Proc. ICASSP 4986-4989, 2010

13 P. Zezula, "Similarity Search: The Metric Space Approach" 32 : 23-38, 2006

14 S. Berrani, "Robust Content-Based Image Searches for Copyright Protection" 70-77, 2003

15 R. Kuhn et, "Rapid Speaker Adaptation in Eigenvoice Space" 8 (8): 695-707, 2000

16 I.T. Jolliffe, "Principal Component Analysis" Springer-Verlag 1986

17 A.K. Noulas, "Online Multimodal Speaker Diarization" 350-357, 2007

18 S. Kullback, "On Information and Sufficiency" 22 (22): 79-86, 1951

19 K. Markov, "Never-Ending Learning with Dynamic Hidden Markov Network" 1437-1440, 2007

20 J. Garofolo, "NIST Rich Transcription 2002 Evaluation: A Preview" 2002

21 A. Dempster, "Maximum Likelihood from Incomplete Data via the EM Algorithm" 39 (39): 1-38, 1977

22 M. Zamalloa, "Low Latency Online Speaker Tracking on the AMI Corpus of Meeting Conversations" 4962-4965, 2010

23 K. Markov, "Improved Novelty Detection for Online GMM Based Speaker Diarization" 363-366, 2008

24 J. Schmalenstroeer, "Fusing Audio and Video Information for Online Speaker Diarization" 1163-1166, 2007

25 L.R. Rabiner, "Fundamentals of Speech Recognition" Prentice-Hall 1993

26 X. Anguera, "Frame Purification for Cluster Comparison in Speaker Diarization" 135-139, 2006

27 K. Chen, "Fast Speaker Adaptation Using Eigenspace-Based Maximum Likelihood Linear Regression" 742-745, 2000

28 T.H. Nguyen, "Cluster Criterion Functions in Spectral Subspace and Their Application in Speaker Clustering" 4085-4088, 2009

29 Mohammad Hossein Moattar, "A Weighted Feature Voting Approach for Robust and Real-Time Voice Activity Detection" 한국전자통신연구원 33 (33): 99-109, 2011

30 B. Mak, "A Study of Various Composite Kernels for Kernel Eigenvoice Speaker Adaptation" 1 : 325-328, 2004

31 C.H. Huang, "A New Eigenvoice Approach to Speaker Adaptation" 109-112, 2004

32 C. Vaquero, "A Hybrid Approach to Online Speaker Diarization" 2638-2631, 2010

33 W. Wang, "A Decision-Tree-Based Online Speaker Clustering" 4477 : 555-562, 2007