Research of a Novel Weak Speech Stream Detection Algorithm

  • Dong-hu Nie
  • Xue-yao Li
  • Ru-bo Zhang
  • Dong Xu
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 4222)


Purpose of speech stream detection is to capture speech stream coming randomly in adverse acoustic environments. A novel robust method for speech stream detection is introduced based on both linear predict code all-pole model and lossless sound tube model to detect speech stream from inputs of wireless speech band communication. It makes use of autocorrelation distribution characteristics of variance sequence of linear predictive residual sequence to formulate two dimensions decision threshold vector. The decision threshold is adaptive to energy of background noise. It can make minimum decisions error. Plenty of signal stream data with various noises under various Signal-to-Noise Ratio and wireless speech band recordings on the spot were used to compare the proposed algorithm respectively with spectrum Entropy and short-time energy algorithm. The experiment results show that the new method for speech stream detection has good detection performance, and it performs well in adverse environments, and the speech stream detected sounds fluently.


Decision Threshold Autocorrelation Coefficient Noisy Speech Endpoint Detection Spectrum Entropy 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Yie, T., Zuoying, W., Dajin, L.: Robust word boundary detection through linear mapping of the sub-band energy in noisy environments. Journal of Tsinghua University (Science and Technology) 42, 953–956 (2002)Google Scholar
  2. 2.
    Li, Q., Zheng, J., Zhou, Q.: A robust, realtime endpoint detector with energy normalization for ASR in adverse environments. In: IEEE International Conference on Acoustics, Speech And Signal Processing (ICASSP 2001), Salt Lake City, USA, vol. 1, pp. 574–577 (2001)Google Scholar
  3. 3.
    Huang, L.S., Yang, C.H.: A novel approach to robust speech endpoint detection in car environments. In: IEEE International Conforence on Acoustics, Speech, and Signal Processing (ICASSP), Istambul, Turkey, pp. 1751–1754 (2003)Google Scholar
  4. 4.
    Jia, C., Xu, B.: An Improved Entropy–based Endpoint Detection Algorithm. In: International Conference on Spoken Language Processing (ICSLP 2002), Taipei, pp. 285–288 (2002)Google Scholar
  5. 5.
    Wang, X., Qi, D., Bingxi, W.: A speech endpoint detector based on eigenspace-energy-entropy. Journal Of China Institute Of Communications 24, 125–132 (2003)Google Scholar
  6. 6.
    Kosmides, E., Dermatas, E., Kokkinakis, G.: Stochastic endpoint detection in noisy speech. In: International Workshop on Speech and Computer (SPECOM 1997), ClujNapoca, Romania, pp. 109–114 (1997)Google Scholar
  7. 7.
    Jie, Z., Xiaodong, W.: Speech Signal Endpoint Detection Method Based on HMM in Noise. Journal of Shanghai Jiaotong University 32, 14–16 (1998)Google Scholar
  8. 8.
    Yuhong, L., Qiao, L., Qiang, R.: Speech signal endpoint detection and separation based on improved fuzzy ART. Systems Engineering and Electronic 26, 1151–1154 (2004)Google Scholar
  9. 9.
    Guangrui, H., Xiaodong, W.: Endpoint Detection of Noisy Speech Based on cepstrum. Acta Electronica Sinica 28, 95–97 (2000)Google Scholar
  10. 10.
    Yaqiang, S., Genliang, F.: Two End Points Detecting and Filtering on Low SNR Speech Signals Based on Short–time Fractal Dimension. Journal of Zhejiang Normal University (Natoral Sciences) 22, 16–21 (1999)Google Scholar
  11. 11.
    Feili, C., Jie, Z.: A New Method of Endpoint Detection Based on Distance of Auto-Correlated Similarity. Journal of Shanghai Jiaotong University 33, 1097–1099 (1999)Google Scholar
  12. 12.
    Qiuan, H., Bo, J., Bingwen, W.: Endpoint detection of Chinese digital speech based on finite state machine. Journal of Hubei University (Natural Science Edition) 26, 35–38 (2004)Google Scholar
  13. 13.
    Javier, R., Jose, C.: Efficient voice activity detection algorithms using long-term speech information. Speech Communication 42, 271–287 (2004)CrossRefGoogle Scholar
  14. 14.
    Liran, S., Xueyao: Speech stream detection based on higher-order statistics. In: International Conference on Machine Learning and Cybernetics, Xi’an, China, vol. 5, pp. 3086–3089 (2003)Google Scholar
  15. 15.
    Xueyao, L., Liran, S.: A new method of robust detection for speech stream. In: Proceedings of 2002 International Conference on Machine Learning and Cybernetics, Beijing, China, vol. 5, pp. 1066–1069 (2002)Google Scholar
  16. 16.
    Liran, S., Xueyao, L.: Speech stream detection based on one and half spectrum. In: Proceedings of the 5th World Congress on Intelligent Control and Automation, Hangzhou, China, vol. 5, pp. 4223–4226 (2004)Google Scholar
  17. 17.
    Rubo, Z., Jiashi, L., Xueyao, L., Liran, S.: Speech stream detection in non-Gaussian background noise based on statistic characteristics of wavelet coefficient. Journal of Harbin Engineering University, Harbin, China 25, 487–490 (2004)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Dong-hu Nie
    • 1
  • Xue-yao Li
    • 1
  • Ru-bo Zhang
    • 1
  • Dong Xu
    • 1
  1. 1.College of Computer Science and TechnologyHarbin Engineering UniversityHarbin CityChina

Personalised recommendations