Détection d'activité vocale basée sur un test d'homogénéité
##plugins.themes.bootstrap3.article.sidebar##
Publié-e :
jun 28, 2014
Mots-clés :
détection d'activité vocale, test d'homogénéité, processus autorégressif.
##plugins.themes.bootstrap3.article.main##
Résumé
Dans cet article, une nouvelle approche pour la détection d'activité vocale (VAD) est proposée. Cette technique est basée sur le test d'homogénéité de deux processus autorégressifs (AR) ; chacun modélise une fenêtre de parole et implique la mesure d'une distance définie. Le test d'homogénéité est formulé comme un test d'hypothèse avec un seuil dérivé analytiquement en fonction d'une probabilité de fausse alarme définie par l'utilisateur. Les résultats utilisant la base de données Aurora montrent l'efficacité de la technique proposée par rapport à d'autres méthodes et normes
##plugins.themes.bootstrap3.article.details##
Comment citer
Rekik, O., & Djeddou, M. (2014). Détection d’activité vocale basée sur un test d’homogénéité. AL-Lisaniyyat, 20(1), 77-85. https://doi.org/10.61850/allj.v20i1.506
Rubrique
Articles
Références
[1] Ramirez J., Gorriz J. M. and Sergura J. C.“Voice activity detction. Fundamentals and speech recognition system robustness in robust Speech recognition and under- standing”.Intech, June 2007. [2] Moattar M. H. and Homayounpour M. M. “A simple but efficient real-time voice activity detection algorithm” 17th European Signal Processing Conference (EUSIPCO 2009).. Clerk Maxwell, A Treatise on Electricity and Magnetism, 3rd ed., vol. 2. Oxford: Clarendon, 1892, Pp.68-73. [3] Wang, Y., Huang, S. and Wei Y., “A voice activity detection algorithm with sub-band detection based on time- frequency characteristics of mandariné , 2013 6th Intemational Congress on Image and Signal Processing (CISP 2013). [4] Savoji M. H., “A robust algorithm for accurate endpointing of speechin: Speech Commun.… 1989,vol. 8, pp. 45-60. [5] Wang J. F. and Chen S. H, “A voice activity detection algorithm based on per- ceptual wavelet packet transform and teager energy — operator,’International Symposium on Chinese Spoken Lan- guage Processing.2002, pp. 177-180. [6] Hung W.W. and Wang H.C., “On the use of weighted filter bank analysis for the derivation of robust MFCC,” IEEE Signal Processing Letters, 2001, pp. 70-73. [7] Chang JH, Kim NS. and Mitra SK. “Voice activity detection based on multiHomogeneity Test based Voice Activity Detection
ple statistical models,” IEEE Trans Signal Processing, 2006, pp.1965-1976. [8] Morales-Cordovilla, JA, Ning Ma, Sanchez, V., Carmona, JL., Peinado,
A.M. and Barker, J., “A pitch based noise estimation technique for robust speech recognition with missing data,” Acous- tics, Speech and Signal Processing (ICASSP), 2011 IEEE International Con- ference on IEEE, 2011, pp. 4808—4811. [9] Rissanen J., “Modeling by shortest data description,” Automatica, vol.14, pp. 465—471, 1978. [10] Levinson N., “The Wiener RMS (Root Mean Square) error Criterion in filter de- sign and prediction,” J Math. Phys., vol. 25, 1947, pp. 261-278. [11] Martinez, R., Gomez, P., and Derouiche, K.,A test of homogeneity for auto- regressive processes ‘’, Int. J. Adapt. Control Signal Process. 2002; 16:213- 242. [12] Hirsch H. and Pearce D., “The Aurora experimental framework for the perfor- mance evaluation of speech recognition systems under noisy conditions.” ISCA ITRW ASR 2000, Paris, France, Septem- ber 18-20. [13] Sohn J., Kim N. S. and Sung W., “A
statistical model-based voice activity de- tection,” IEEE Signal Process. Lett., vol. 16, no. 1, pp. 1-3, Jan. 1999. [14] Benyassine A., Shlomot E..Su H.-Y.
Massaloux, D, Lamblin C. and Petit J-P., “Ttu-t recommendation g.729 annex b : A
silence compression scheme for use with g.729 optimized for v.70 digital simulta- neous voice and data applications. IEEE Communications Magazine, 35(9):64-73, September 1997. [15] European Standard (Telecommunications series). Voice activity detector (vad) for adaptive multi-rate (amr) speech trafic channels, 1999. ETSI EN 301 708 v7.11 standard description. [16] European standard (Telecomunications- nseries). Transmission and quality aspect
front-end feature extarction algorithm, 2007. ETSI ES 202050 standard descriprtion.
ple statistical models,” IEEE Trans Signal Processing, 2006, pp.1965-1976. [8] Morales-Cordovilla, JA, Ning Ma, Sanchez, V., Carmona, JL., Peinado,
A.M. and Barker, J., “A pitch based noise estimation technique for robust speech recognition with missing data,” Acous- tics, Speech and Signal Processing (ICASSP), 2011 IEEE International Con- ference on IEEE, 2011, pp. 4808—4811. [9] Rissanen J., “Modeling by shortest data description,” Automatica, vol.14, pp. 465—471, 1978. [10] Levinson N., “The Wiener RMS (Root Mean Square) error Criterion in filter de- sign and prediction,” J Math. Phys., vol. 25, 1947, pp. 261-278. [11] Martinez, R., Gomez, P., and Derouiche, K.,A test of homogeneity for auto- regressive processes ‘’, Int. J. Adapt. Control Signal Process. 2002; 16:213- 242. [12] Hirsch H. and Pearce D., “The Aurora experimental framework for the perfor- mance evaluation of speech recognition systems under noisy conditions.” ISCA ITRW ASR 2000, Paris, France, Septem- ber 18-20. [13] Sohn J., Kim N. S. and Sung W., “A
statistical model-based voice activity de- tection,” IEEE Signal Process. Lett., vol. 16, no. 1, pp. 1-3, Jan. 1999. [14] Benyassine A., Shlomot E..Su H.-Y.
Massaloux, D, Lamblin C. and Petit J-P., “Ttu-t recommendation g.729 annex b : A
silence compression scheme for use with g.729 optimized for v.70 digital simulta- neous voice and data applications. IEEE Communications Magazine, 35(9):64-73, September 1997. [15] European Standard (Telecommunications series). Voice activity detector (vad) for adaptive multi-rate (amr) speech trafic channels, 1999. ETSI EN 301 708 v7.11 standard description. [16] European standard (Telecomunications- nseries). Transmission and quality aspect
front-end feature extarction algorithm, 2007. ETSI ES 202050 standard descriprtion.