Subject: Re: How to classify/categorize sounds From: Holger Prante <Prante(at)t-online.de> Date: Sun, 25 Mar 2001 20:48:26 +0200This is a multi-part message in MIME format. ------=_NextPart_000_0016_01C0B56C.F064B0C0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Pawel, we made binaural recordings of 25 environmental sounds, which were then played back to a jury via headphones. The jury had the task to judge = upon the sounds by means of several semantic differentials. By factor = analysis of the subject's answers 6 invariant principal dimensions were extracted, = which can be used to differenciate between the sounds. Whereas the above analysis accounts for the description of the = judgements, however, I trained neural networks to do the same task, namely to map = the sound recordings onto the dimensions extracted from the test. The = temporal extension to the self-organized-feature-map produced the best results: = the sounds were clustered according to their spectral and short-term = temporal features. This way, sounds with tonal, noisy, stationary or fluctuating features were arranged on a two dimensional map. The work is described in my Ph.D.-thesis, however you can find the = feature map results also in Prante, Holger U. & Koop, Lars; Classification of Sounds with Temporal Feature Maps. In: Schick A., Meis, M., Reckhardt, C. (eds.) 2000, Contributions to Psychological Acoustics. Results of the 8th Oldenburg Symposium on Psychological Acoustics. Oldenburg University Press. Regards, Holger -------------------- Holger U. Prante Institute of Technical Acoutics Technical University of Berlin pran2133(at)mailszrz.zrz.tu-berlin.de -----Urspr=FCngliche Nachricht----- Von: Pawel Kusmierek <pq(at)NENCKI.GOV.PL> An: AUDITORY(at)LISTS.MCGILL.CA <AUDITORY(at)LISTS.MCGILL.CA> Datum: 20 March, 2001 2:34 PM Betreff: How to classify/categorize sounds I run animal experiments which involve sound recognition and/ordiscrimination. Many various sound stimuli are used (e.g., tones, instruments, animal sounds (including birds), machines, artficial sounds, knocks, bells etc.). Is it possible to describe quantitatively such various sounds (along multiple dimensions, I suppose) in order to find quantitative indices of dfference, either in multidimensional space, or along a single dimension? Any help would be greatly appreciated. Pawel Kusmierek ************************************* Pawel Kusmierek Department of Neurophysiology Nencki Institute of Experimental Biology 3, Pasteur St., 02-093 Warsaw, Poland tel. (48-22) 659 85 71 ex 379 fax (48-22) 822 53 42 E-mail pq(at)nencki.gov.pl Or: kusmierek(at)yahoo.com kusmierek(at)poczta.arena.pl ICQ 11740175 =20 =20 ------=_NextPart_000_0016_01C0B56C.F064B0C0 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable <!DOCTYPE HTML PUBLIC "-//W3C//DTD W3 HTML//EN"> <HTML> <HEAD> <META content=3Dtext/html;charset=3Diso-8859-1 = http-equiv=3DContent-Type> <META content=3D'"MSHTML 4.72.3110.7"' name=3DGENERATOR> </HEAD> <BODY bgColor=3D#f2ecd7> <DIV> <DIV><FONT size=3D2>Pawel,<BR><BR>we made binaural recordings of 25 = environmental=20 sounds, which were then<BR>played back to a jury via headphones. The = jury had=20 the task to judge upon<BR>the sounds by means of several semantic = differentials.=20 By factor analysis of<BR>the subject's answers 6 invariant principal = dimensions=20 were extracted, which<BR>can be used to differenciate between the=20 sounds.<BR>Whereas the above analysis accounts for the description of = the=20 judgements,<BR>however, I trained neural networks to do the same task, = namely to=20 map the<BR>sound recordings onto the dimensions extracted from the test. = The=20 temporal<BR>extension to the self-organized-feature-map produced the = best=20 results: the<BR>sounds were clustered according to their spectral and = short-term=20 temporal<BR>features. This way, sounds with tonal, noisy, stationary or=20 fluctuating<BR>features were arranged on a two dimensional map.<BR>The = work is=20 described in my Ph.D.-thesis, however you can find the feature<BR>map = results=20 also in<BR>Prante, Holger U. & Koop, Lars; Classification of Sounds = with=20 Temporal<BR>Feature Maps. In: Schick A., Meis, M., Reckhardt, C. (eds.)=20 2000,<BR>Contributions to Psychological Acoustics. Results of the 8th=20 Oldenburg<BR>Symposium on Psychological Acoustics. Oldenburg University=20 Press.<BR><BR>Regards,<BR>Holger<BR><BR>--------------------<BR>Holger = U.=20 Prante<BR>Institute of Technical Acoutics<BR>Technical University of=20 Berlin<BR><A=20 href=3D"mailto:pran2133(at)mailszrz.zrz.tu-berlin.de">pran2133(at)mailszrz.zrz.= tu-berlin.de</A><BR><BR><BR>-----Ursprüngliche=20 Nachricht-----<BR>Von: Pawel Kusmierek <<A=20 href=3D"mailto:pq(at)NENCKI.GOV.PL">pq(at)NENCKI.GOV.PL</A>><BR>An: <A=20 href=3D"mailto:AUDITORY(at)LISTS.MCGILL.CA">AUDITORY(at)LISTS.MCGILL.CA</A> = <<A=20 href=3D"mailto:AUDITORY(at)LISTS.MCGILL.CA">AUDITORY(at)LISTS.MCGILL.CA</A>>= <BR>Datum:=20 20 March, 2001 2:34 PM<BR>Betreff: How to classify/categorize=20 sounds<BR><BR><BR>I run animal experiments which involve sound=20 recognition<BR>and/ordiscrimination. Many various sound stimuli = are used=20 (e.g.,<BR>tones, instruments, animal sounds (including birds),=20 machines,<BR>artficial sounds, knocks, bells etc.). Is it possible = to=20 describe<BR>quantitatively such various sounds (along multiple = dimensions,=20 I<BR>suppose) in order to find quantitative indices of dfference, either = in<BR>multidimensional space, or along a single dimension?<BR><BR>Any = help would=20 be greatly appreciated.<BR><BR>Pawel=20 Kusmierek<BR><BR><BR>*************************************<BR><BR>Pawel=20 Kusmierek<BR>Department of Neurophysiology<BR>Nencki Institute of = Experimental=20 Biology<BR>3, Pasteur St., 02-093 Warsaw, Poland<BR><BR>tel. (48-22) 659 = 85 71=20 ex 379<BR>fax (48-22) 822 53 42<BR>E-mail <A=20 href=3D"mailto:pq(at)nencki.gov.pl">pq(at)nencki.gov.pl</A><BR>Or:<BR><A=20 href=3D"mailto:kusmierek(at)yahoo.com">kusmierek(at)yahoo.com</A><BR><A=20 href=3D"mailto:kusmierek(at)poczta.arena.pl">kusmierek(at)poczta.arena.pl</A><B= R>ICQ=20 11740175<BR></FONT></DIV> <DIV><FONT size=3D2></FONT> </DIV> <DIV><FONT size=3D2></FONT> </DIV></DIV> <DIV> </DIV></BODY></HTML> ------=_NextPart_000_0016_01C0B56C.F064B0C0--