This Annex A replaces the previous Annex A to ITU-T Rec. P.862 (February 2001), which defines the Perceptual Evaluation of Speech Quality (PESQ) model for objective speech quality measurement for telephone networks and speech codecs. It clarifies the sampling rate that must be offered by implementations of PESQ, and sets out an extended set of conformance tests that support this.

It is not anticipated that this Annex A will affect any existing implementations of PESQ. Future implementations of PESQ that only offer 8 kHz sampling rate are accepted as valid and must meet the new conformance test 1(b) proposed below. Future implementations at 16 kHz sampling rate must also meet the new conformance test 2(a).

The ANSI-C reference implementation of PESQ remains unchanged.

The only files that are modified relative to ITU-T Rec. P.862 (02/2001) are:

supp23_16k.txt, supp23_16k.bat, supp23_8k.txt, supp23_8k.bat;

voipref_16k.txt, voipref_16k.bat, voipref_8k.txt, voipref_8k.bat;


These files are available for free on ITU Publications website at, under P.862 as "P.862 (2001) Amendment 1".