Telephone Speech Endpoint Detection using Mean-Delta Feature

Ouzounov, Atanas

doi:10.2478/cait-2014-0025

Abstract

In the study the efficiency of three features for trajectory-based endpoint detection is experimentally evaluated in the fixed-text Dynamic Time Warping (DTW) - a based speaker verification task with short phrases of telephone speech. The employed features are Modified Teager Energy (MTE), Energy-Entropy (EE) feature and Mean-Delta (MD) feature. The utterance boundaries in the endpoint detector are provided by means of state automaton and a set of thresholds based only on trajectory characteristics. The training and testing have been done with noisy telephone speech (short phrases in Bulgarian language with length of about 2 s) selected from BG-SRDat corpus. The results of the experiments have shown that the MD feature demonstrates the best performance in the endpoint detection tests in terms of the verification rate.

References

1. Bengio, S., J. Mariethoz. A Statistical Significance Test for Person Authentication, ODYSSEY. - In: The Speaker and Language Recognition Workshop, 2004, 237-244.
Search in Google Scholar
2. Burileanu, C., D. Moraru, L. Bojan, M. Puchiu, A. Stan. On Performance Improvement of a Speaker Verification System Using Vector Quantization, Cohorts and Hybrid Cohort-World Models. - International Journal of Speech Technology, 2002, No 5, 247-257.10.1023/A:1020244924468
Search in Google Scholar
3. Gerven, S., F. Xie. A Comparative Study of Speech Detection Methods. Eurospeech, 1997, 1095-1098.10.21437/Eurospeech.1997-108
Search in Google Scholar
4. Gu, L., S. Zahorian. A New Robust Algorithm for Isolated Word Endpoint Detection. - IEEE ICASSP, Vol. IV, 2002, 4161-4164.10.1109/ICASSP.2002.5745582
Search in Google Scholar
5. Huang, L., C. Yang. A Novel Approach to Robust Speech Endpoint Detection in Car Environment. - In: IEEE ICASSP, 2000, 1751-1754.
Search in Google Scholar
6. J i a, C., B. X u. An Improved Entropy Based Endpoint Detection Algorithm. - ISCSLP, 2002, 96100.
Search in Google Scholar
7. Li, Q., J. Zheng, A. Tsai, Q. Zhou. Robust Endpoint Detection and Energy Normalization for Real-Time Speech and Speaker Recognition. - IEEE Transaction on SAP, Vol. 10, 2002, No 3, 146-157.10.1109/TSA.2002.1001979
Search in Google Scholar
8. Fawcett, T. An Introduction to ROC Analysis. - Pattern Recognition Letters, Vol. 27, 2006, No 8, 861-874.10.1016/j.patrec.2005.10.010
Search in Google Scholar
9. Mesa-Navarro, J., A. Moreno-Bilbao, E. Lleida-Solano. An Improved Speech Endpoint Detection System in Noisy Environments by Means of Third-Order Spectra. -IEEE Signal Processing Letters, Vol. 6, 1999, No 9, 224-226.10.1109/97.782065
Search in Google Scholar
10. Myers, C., L. Rabiner, A. Rosenberg. Performance Tradeoffs in Dynamic Time Warping Algorithms for Isolated Word Recognition. - IEEE Transactions on ASSP, Vol. 28, 1980, No 6, 623-635.10.1109/TASSP.1980.1163491
Search in Google Scholar
11. McAulay, R., M. Malpass. Speech Enhancement Using a Soft-Decision Noise Suppression Filter. - IEEE Transactions on ASSP, Vol. 28, 1980, No 2, 137-145.10.1109/TASSP.1980.1163394
Search in Google Scholar
12. Ouzounov, A. Endpoint Detection Algorithm. - In: VoicePass - Real-Time Speaker Verification System Based on DSP Board Linkon FC-3000. Internal Report AdVoice, Ltd., 1996.
Search in Google Scholar
13. Ouzounov, A. BG-SRDat: A Corpus in Bulgarian Language for Speaker Recognition over Telephone Channels. - Cybernetics and Information Technologies, Vol. 3, 2003, No 2, 101-108.
Search in Google Scholar
14. Ouzounov, A. A Robust Feature for Speech Detection. - Cybernetics and Information Technologies, Vol. 4, 2004, No 2, 3-14.
Search in Google Scholar
15. Ouzounov, A. Robust Features and Neural Network for Noisy Speech Detection. - Cybernetics and Information Technologies, Vol. 6, 2006, No 3, 75-84.
Search in Google Scholar
16. Ouzounov, A. Cepstral Features and Text-Dependent Speaker Identification - A Comparative Study. - Cybernetics and Information Technologies, Vol. 10, 2010, No 1, 1-12.
Search in Google Scholar
17. Ramirez, J., J. Segura, C. Benitez, A. De la Torre, A. Rubio. Efficient Voice Activity Detection Algorithms Using Long-Term Speech Information. - Speech Communication, Vol. 42, 2004, No 3-4, 271-287.10.1016/j.specom.2003.10.002
Search in Google Scholar
18. Ramirez, J., P. Yelamos, J. Gorriz, J. Seguraet. SVM-Based Speech Endpoint Detection Using Contextual Speech Features. - Electronics Letters, Vol. 42, 2006, No 7, 426-428.10.1049/el:20064068
Search in Google Scholar
19. Seok, J., K. Bae. A Novel Endpoint Detection Using Discrete Wavelet Transform. - IEICE Transaction on Inf. & Syst., Vol. E82-D, 1999, No 11, 1489-1491.
Search in Google Scholar
20. Shin, W., B. Lee, Y. Lee, J. Lee. Speech/Non-Speech Classification Using Multiple Features for Robust Endpoint Detection. - IEEE ICASSP, 2000, 1399-1402.
Search in Google Scholar
21. Wu, B. F., K. C. Wang. Robust Endpoint Detection Algorithm Based on the Adaptive Band-Partitioning Spectral Entropy in Adverse Environments. - IEEE Transactions on SAP, Vol. 13, 2005, No 5, 762-775.10.1109/TSA.2005.851909
Search in Google Scholar
22. Yamamoto K., F. Jabloun, K. Reinhard, A. Kawamura. Robust Endpoint Detection for Speech Recognition Based on Discriminative Feature Extraction. - In: IEEE ICASSP, Vol. I, 2006, 805-808.
Search in Google Scholar
23. Zelinski, R., F. Class. A Learning Procedure for Speaker-Dependent Word Recognition System Based on Sequential Processing of Input Tokens. - In: IEEE ICASSP, 1983, 1053-1056.
Search in Google Scholar
24. Zhao H., L. Zhao, K. Zhao, G. Wang. Voice Activity Detection Based on Distance Entropy in Noisy Environment. - In: Fifth International Joint Conference on INC, IMS and IDC, 2009, 1364-1367.10.1109/NCM.2009.134
Search in Google Scholar
25. Zhang, Z., S. Furui. Noisy Speech Recognition Based on Robust End-Point Detection and Model Adaptation. - In: IEEE ICASSP, Vol. 1, 2005, 441-444.
Search in Google Scholar
26. Center for Spoken Language Understanding, Speech Enhancement and Assessment Resource (SpEAR) Database. Oregon Graduate Institute of Science and Technology. http://www.cslu.ogi.edu/nsel/data/SpEAR_lombard.html
Search in Google Scholar

Telephone Speech Endpoint Detection using Mean-Delta Feature

Abstract

Paradigm

My account