Re: AUDITORY Digest - 26 Sep 2010 to 27 Sep 2010 - Special issue (#2010-228) (karan behar )


Subject: Re: AUDITORY Digest - 26 Sep 2010 to 27 Sep 2010 - Special issue (#2010-228)
From:    karan behar  <karanbehar@xxxxxxxx>
Date:    Wed, 6 Oct 2010 12:57:40 -0400
List-Archive:<http://lists.mcgill.ca/scripts/wa.exe?LIST=AUDITORY>

--000e0cdf9dfc1df4ac0491f5ae23 Content-Type: text/plain; charset=ISO-8859-1 Dear Sir Greetings. I apologize for asking mundane and trite questions in my previous posts. Thank you very much for your suggestions. I wanted to ask a question on Mel Frequency Cepstral Coefficients for obtaining feature vectors. In many papers I have read about people obtaining 13 coefficients and in some about obtaining 12 MFCC using FFT and then using them to treat the network. The example of such paper is Logan B. "Mel frequency cepstral coefficients for music modeling<http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.11.9216&rep=rep1&type=pdf> ", International Symposium of Music Information and for 12 coefficients Fukada T. "An adaptive algorithm for mel-*cepstral *analysis of speech<http://www.sp.nitech.ac.jp/%7Etokuda/selected_pub/pdf/conference/fukada_icassp1992.pdf>", ICASSP, 1992. 1. Why do we need 13 or 12 coefficients. What are the differences between them. 2. I am having hard time to understand the what each of the thirteen coefficients physically and mathematically represent for MFCC 13 and 12 coefficients. For example if first coefficient represents frequency and second represents timber etc. I have been using 13 MFCC to train my network but I do not know everything that works behind using 13 or 12 coefficients. Through and fundamental papers addressing the mathematics and significance of these questions will be very helpful. Please advise. Regards Karan --000e0cdf9dfc1df4ac0491f5ae23 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Dear Sir<br> <br> Greetings. I apologize for asking mundane and trite questions in my previou= s posts. Thank you very much for your suggestions. <br> <br> I wanted to ask a question on Mel Frequency <span style=3D"background: none= repeat scroll 0% 0% yellow;" class=3D"J-JK9eJ-PJVNOc">Cepstral</span> Coefficients for obtaining feature vectors. In many papers I have read=20 about people obtaining 13 coefficients and in some about obtaining 12 <span= style=3D"background: none repeat scroll 0% 0% yellow;" class=3D"J-JK9eJ-PJ= VNOc">MFCC</span> using <span style=3D"background: none repeat scroll 0% 0%= yellow;" class=3D"J-JK9eJ-PJVNOc">FFT</span> and then using them to treat = the network. The example of such paper is Logan B. <span style=3D"color: rg= b(0, 0, 0);">&quot;<a style=3D"font-weight: normal;" href=3D"http://citesee= rx.ist.psu.edu/viewdoc/download?doi=3D10.1.1.11.9216&amp;rep=3Drep1&amp;typ= e=3Dpdf" class=3D"yC0">Mel frequency <span style=3D"background: none repeat= scroll 0% 0% yellow;" class=3D"J-JK9eJ-PJVNOc">cepstral</span> coefficient= s for music modeling</a><span style=3D"font-weight: normal;">&quot;</span><= /span>, International Symposium of Music Information and for 12 coefficient= s <span style=3D"background: none repeat scroll 0% 0% yellow;" class=3D"J-J= K9eJ-PJVNOc">Fukada</span> T.=A0 &quot;<a href=3D"http://www.sp.nitech.ac.j= p/%7Etokuda/selected_pub/pdf/conference/fukada_icassp1992.pdf" class=3D"yC6= ">An adaptive algorithm for <span style=3D"background: none repeat scroll 0= % 0% yellow;" class=3D"J-JK9eJ-PJVNOc">mel</span>-<b><span style=3D"backgro= und: none repeat scroll 0% 0% yellow;" class=3D"J-JK9eJ-PJVNOc">cepstral</s= pan> </b>analysis of speech</a>&quot;, <span style=3D"background: none repe= at scroll 0% 0% yellow;" class=3D"J-JK9eJ-PJVNOc">ICASSP</span>, 1992. <br> <br> 1. Why do we need 13 or 12 coefficients. What are the differences between t= hem. <br> <br> 2. I am having hard time to understand the what each of the thirteen coeffi= cients physically and mathematically=A0 represent for <span style=3D"backgr= ound: none repeat scroll 0% 0% yellow;" class=3D"J-JK9eJ-PJVNOc">MFCC</span= > 13 and 12 coefficients. For example if first coefficient represents frequ= ency and second represents timber etc. <br> <br>I have been using 13 MFCC to train my network but I do not know everyth= ing that works behind using 13 or 12 coefficients.<br><br>Through and funda= mental papers addressing the mathematics and significance of these question= s will be very helpful.<br> <br> Please advise.<br> <br> Regards<br> <br> <span style=3D"background: none repeat scroll 0% 0% yellow;" class=3D"J-JK9= eJ-PJVNOc">Karan</span>=20 --000e0cdf9dfc1df4ac0491f5ae23--


This message came from the mail archive
/home/empire6/dpwe/public_html/postings/2010/
maintained by:
DAn Ellis <dpwe@ee.columbia.edu>
Electrical Engineering Dept., Columbia University