V_DISTITPF calculates the Itakura spectral distance between power spectra D=(PF1,PF2,MODE) Inputs: PF1,PF2 Power spectra to be compared. Each row represents a power spectrum: the first and last columns represent the DC and Nyquist terms respectively. PF1 and PF2 must have the same number of columns. MODE Character string selecting the following options: 'x' Calculate the full distance matrix from every row of PF1 to every row of PF2 'd' Calculate only the distance between corresponding rows of PF1 and PF2 The default is 'd' if PF1 and PF2 have the same number of rows otherwise 'x'. Output: D If MODE='d' then D is a column vector with the same number of rows as the shorter of PF1 and PF2. If MODE='x' then D is a matrix with the same number of rows as PF1 and the same number of columns as PF2'. If ave() denotes the average over +ve and -ve frequency, the Itakura spectral distance is log(ave(pf1/pf2)) - ave(log(pf1/pf2)) The Itakura distance is gain-independent, i.e. v_distitpf(g*pf1,pf2) is independent of g.
0001 function d=v_distitpf(pf1,pf2,mode) 0002 %V_DISTITPF calculates the Itakura spectral distance between power spectra D=(PF1,PF2,MODE) 0003 % 0004 % Inputs: PF1,PF2 Power spectra to be compared. Each row represents a power spectrum: the first 0005 % and last columns represent the DC and Nyquist terms respectively. 0006 % PF1 and PF2 must have the same number of columns. 0007 % 0008 % MODE Character string selecting the following options: 0009 % 'x' Calculate the full distance matrix from every row of PF1 to every row of PF2 0010 % 'd' Calculate only the distance between corresponding rows of PF1 and PF2 0011 % The default is 'd' if PF1 and PF2 have the same number of rows otherwise 'x'. 0012 % 0013 % Output: D If MODE='d' then D is a column vector with the same number of rows as the shorter of PF1 and PF2. 0014 % If MODE='x' then D is a matrix with the same number of rows as PF1 and the same number of columns as PF2'. 0015 % 0016 % If ave() denotes the average over +ve and -ve frequency, the Itakura spectral distance is 0017 % 0018 % log(ave(pf1/pf2)) - ave(log(pf1/pf2)) 0019 % 0020 % The Itakura distance is gain-independent, i.e. v_distitpf(g*pf1,pf2) is independent of g. 0021 0022 % The Itakura distance can also be calculated directly from AR coefficients; providing np is large 0023 % enough, the values of d0 and d1 in the following will be very similar: 0024 % 0025 % np=255; d0=v_distitar(ar1,ar2); d1=v_distitpf(v_lpcar2pf(ar1,np),v_lpcar2pf(ar2,np)) 0026 % 0027 0028 % Ref: A.H.Gray Jr and J.D.Markel, "Distance measures for speech processing", IEEE ASSP-24(5): 380-391, Oct 1976 0029 % L. Rabiner abd B-H Juang, "Fundamentals of Speech Recognition", Section 4.5, Prentice-Hall 1993, ISBN 0-13-015157-2 0030 % F. Itakura, "Minimum prediction residual principle applied to speech recognition", IEEE ASSP-23: 62-72, 1975 0031 0032 0033 % Copyright (C) Mike Brookes 1997 0034 % Version: $Id: v_distitpf.m 10865 2018-09-21 17:22:45Z dmb $ 0035 % 0036 % VOICEBOX is a MATLAB toolbox for speech processing. 0037 % Home page: http://www.ee.ic.ac.uk/hp/staff/dmb/voicebox/voicebox.html 0038 % 0039 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% 0040 % This program is free software; you can redistribute it and/or modify 0041 % it under the terms of the GNU General Public License as published by 0042 % the Free Software Foundation; either version 2 of the License, or 0043 % (at your option) any later version. 0044 % 0045 % This program is distributed in the hope that it will be useful, 0046 % but WITHOUT ANY WARRANTY; without even the implied warranty of 0047 % MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the 0048 % GNU General Public License for more details. 0049 % 0050 % You can obtain a copy of the GNU General Public License from 0051 % http://www.gnu.org/copyleft/gpl.html or by writing to 0052 % Free Software Foundation, Inc.,675 Mass Ave, Cambridge, MA 02139, USA. 0053 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% 0054 0055 [nf1,p2]=size(pf1); 0056 p1=p2-1; 0057 nf2=size(pf2,1); 0058 if nargin<3 | isempty(mode) mode='0'; end 0059 if any(mode=='d') | (mode~='x' & nf1==nf2) 0060 nx=min(nf1,nf2); 0061 r=pf1(1:nx,:)./pf2(1:nx,:); 0062 q=log(r); 0063 d=log((sum(r(:,2:p1),2)+0.5*(r(:,1)+r(:,p2)))/p1)-(sum(q(:,2:p1),2)+0.5*(q(:,1)+q(:,p2)))/p1; 0064 else 0065 r=permute(pf1(:,:,ones(1,nf2)),[1 3 2])./permute(pf2(:,:,ones(1,nf1)),[3 1 2]); 0066 q=log(r); 0067 d=log((sum(r(:,:,2:p1),3)+0.5*(r(:,:,1)+r(:,:,p2)))/p1)-(sum(q(:,:,2:p1),3)+0.5*(q(:,:,1)+q(:,:,p2)))/p1; 0068 end