New method vs. PESQ for perceptural voice quality testing.
What are advantages of a well known standardized method in comparison with a noname approach? Publicity is the main advantage! In modern World one creates demand and fits his own supply into it, but we do believe that the World is small enough to find new technologies, to give the a hard test and if they quality take their niche in business success.
What are potential advantages of a noname method if any? We believe they do exist. They are flexibility, performance, reasonable advantages, new vision… so, why don’t we see what the new method has to say against P.862, the giant of voice quality evaluation technology:
1. In PESQ test signals must be speech-like because many systems are optimized for speech and respond in an unrepresentative way to non-speech signals (e.g. tones, noise, ITU-T P.50).
- the new method claims that it works for non-speech signals also, so basically one can test different spectrums, or if just in case we get another frequency (out of speech band) in the channel that will not affect our system.
2. PESQ demands test signal is set by tester and thus vendor estimations may vary from end customer estimations.
- the new approach doesn’t demand that
3. PESQ works with speech only
- the new approach is suitable for testing any audio in principle
4. PESQ performs signal level equalization what theoretically is not that good because when speaking different sound volumes may have different spectrums.
5. PESQ has the so-called “disturbance processing”
- the new method does not have it and if signal spectrum is disturbed that will be reflected in the estimation by obviously lowing it down
