This implementation of Matlab was originally taken from https://ecs.utdallas.edu/loizou/speech/software.htm
Octave must be installed
apt install octave
pip install oct2py Cython # install Cython before installing pesq
pip install pesqUsage:
[Csig,Cbak,Covl,segSNR]=composite(cleanfile.wav,enhanced.wav)where Csig is the predicted rating of speech distortion, Cbak is the predicted rating of background distortion, Covl is the predicted rating of overall quality.
You may run example files included in the zip file. In MATLAB, type:
>> [c,b,o,s]=composite('sp09.wav','enhanced_logmmse.wav')
c = 3.3050
b = 2.6160
o = 2.7133
s = 3.9917where sp09.wav is the clean file and enhanced_logmmse.wav is the enhanced file.
MATLAB Source: https://ecs.utdallas.edu/loizou/speech/software.htm
@ARTICLE{4389058,
author={Y. {Hu} and P. C. {Loizou}},
journal={IEEE Transactions on Audio, Speech, and Language Processing},
title={Evaluation of Objective Quality Measures for Speech Enhancement},
year={2008},
volume={16},
number={1},
pages={229-238},
}