Voice Activity Detection Based on Non-negative Matrix Factorization 


Vol. 35,  No. 8, pp. 661-666, Aug.  2010


PDF
  Abstract

In this paper, we apply a likelihood ratio test (LRT) to a non-negative matrix factorization (NMF) based voice activity detection (VAD) to find optimal threshold. In our approach, the NMF based VAD is expressed as Euclidean distance between noise basis vector and input basis vector which are extracted through NMF. The optimal threshold each of noise environments depend on NMF results distribution in noise region which is estimated statistical model-based VAD. According to the experimental results, the proposed approach is found to be effective for statistical model-based VAD using LRT.

  Statistics
Cumulative Counts from November, 2022
Multiple requests among the same browser session are counted as one view. If you mouse over a chart, the values of data points will be shown.


  Cite this article

[IEEE Style]

S. Kang and J. Chang, "Voice Activity Detection Based on Non-negative Matrix Factorization," The Journal of Korean Institute of Communications and Information Sciences, vol. 35, no. 8, pp. 661-666, 2010. DOI: .

[ACM Style]

Sang-Ick Kang and Joon-Hyuk Chang. 2010. Voice Activity Detection Based on Non-negative Matrix Factorization. The Journal of Korean Institute of Communications and Information Sciences, 35, 8, (2010), 661-666. DOI: .

[KICS Style]

Sang-Ick Kang and Joon-Hyuk Chang, "Voice Activity Detection Based on Non-negative Matrix Factorization," The Journal of Korean Institute of Communications and Information Sciences, vol. 35, no. 8, pp. 661-666, 8. 2010.

Search
(IN TITLE, AUTHOR, ABSTRACT,KEYWORDS)

Advanced Search ▼

POPULAR KEYWORDS
(TOP 10 KEYWORDS)

Recent Publications
(LAST 3 YEARS)






pISSN : 1226 - 4717
eISSN : 2287 - 3880