Noise Subspace Fuzzy C-means Clustering for Robust Speech Re(10)

Abstract. In this paper a fuzzy C-means (FCM) based approach for speech/non-speech discrimination is developed to build an effective voice activity detection (VAD) algorithm. The proposed VAD method is based on a soft-decision clustering approach built ove

> > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > >Your camera-ready and source files are due on February 21, 2006. The same date expires also the Registration to ICCSA 2006. If none of the authors have registered and paid by the registration deadline the paper WILL NOT be included on the special issue of LNCS. Only 1 oral presentation per person is included in the Conference fee. If an author have to show up more than 1 paper, the "Additional Paper presentation" fee is due. For presenters paying the student rate, the 2nd oral presentation will be due the full student rate. Please fill the Copyright Tranfer Form available at the URL: /sgw/cda/pageitems/document/cda_downloadd ocument/0,11996,0-0-45-72914-0,00.pdf Details for filling the Copyright Transfer Form will be available at the Conference web site: Congratulations on having your paper accepted. We look forward to seeing you in Glasgow. Best Regards, Osvaldo Gervasi ICCSA 2006 Program Chair ----------------------oooooooooooooooooooooo---------------------Osvaldo Gervasi - High Performance Computing Laboratory - Dept. of Mathematics and Computer Science, University of Perugia, Address: Via Vanvitelli 1; I-06123 Perugia (Italy) Voice: +39 0755855048 Facsim.: +39 0755855024 Mobil:+39 3485109938 E-mail: ogervasi@ http://ogervasi.unipg.it ==================================================================== REFEREE MESSAGES: ==================================================================== *=--=*=--=*=--=*=--=*=--=*=--=*=--=*=--=*=--=*=--=*=--=*=--=*=--=*=--= *=--=*= First reviewer's review: >>> Readability <<< Acceptable>>> Originality <<< Very good>>> Significance <<< Acceptable>>> Relevance <<<2