v_stoi2prob

V_STOI2PROB convert STOI to probability

  Inputs:  S(M,N)       matrix containing STOI values
           M            mapping: 'i' IEEE sentences [default]
                                 'd' Dantale corpus

 Outputs:  P(M,N)       Corresponding probability values

 STOI is an intelligibility metric described in [1]. The
 mapping from STOI to intelligibilty is corpus-dependent:
 this functions implements two mappings given in [1].

 [1]    C. H. Taal, R. C. Hendriks, R. Heusdens, and J. Jensen.
       An algorithm for intelligibility prediction of time-frequency weighted noisy speech.
       IEEE Trans. Audio, Speech, Language Processing, 19 (7): 2125-2136, 2011.
       doi: 10.1109/TASL.2011.2114881.

0001 function p=v_stoi2prob(s,m)
0002 %V_STOI2PROB convert STOI to probability
0003 %
0004 %  Inputs:  S(M,N)       matrix containing STOI values
0005 %           M            mapping: 'i' IEEE sentences [default]
0006 %                                 'd' Dantale corpus
0007 %
0008 % Outputs:  P(M,N)       Corresponding probability values
0009 %
0010 % STOI is an intelligibility metric described in [1]. The
0011 % mapping from STOI to intelligibilty is corpus-dependent:
0012 % this functions implements two mappings given in [1].
0013 %
0014 % [1]    C. H. Taal, R. C. Hendriks, R. Heusdens, and J. Jensen.
0015 %       An algorithm for intelligibility prediction of time-frequency weighted noisy speech.
0016 %       IEEE Trans. Audio, Speech, Language Processing, 19 (7): 2125-2136, 2011.
0017 %       doi: 10.1109/TASL.2011.2114881.
0018 
0019 %      Copyright (C) Mike Brookes 2014
0020 %      Version: $Id: v_stoi2prob.m 10865 2018-09-21 17:22:45Z dmb $
0021 %
0022 %   VOICEBOX is a MATLAB toolbox for speech processing.
0023 %   Home page: http://www.ee.ic.ac.uk/hp/staff/dmb/voicebox/voicebox.html
0024 %
0025 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
0026 %   This program is free software; you can redistribute it and/or modify
0027 %   it under the terms of the GNU General Public License as published by
0028 %   the Free Software Foundation; either version 2 of the License, or
0029 %   (at your option) any later version.
0030 %
0031 %   This program is distributed in the hope that it will be useful,
0032 %   but WITHOUT ANY WARRANTY; without even the implied warranty of
0033 %   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
0034 %   GNU General Public License for more details.
0035 %
0036 %   You can obtain a copy of the GNU General Public License from
0037 %   http://www.gnu.org/copyleft/gpl.html or by writing to
0038 %   Free Software Foundation, Inc.,675 Mass Ave, Cambridge, MA 02139, USA.
0039 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
0040 
0041 if nargin<2 || ~numel(m) || m(1)=='i'
0042     a=-17.4906;
0043     b=9.6921;
0044     if ~nargin
0045         s=[];
0046     end
0047 else
0048     a=-14.5435;
0049     b=7.0792;
0050 end
0051 p=1./(1+exp(a*s+b));
0052 if ~nargout
0053     if ~nargin
0054         s=linspace(0,1,100);
0055     end
0056     plot(s,v_stoi2prob(s,'d'),':k',s,v_stoi2prob(s,'i'),'-k')
0057     xlabel('STOI');
0058     ylabel('Intelligibility');
0059     legend('Dantale corpus','IEEE sentences','location','best');
0060 end