Recommendation ITU-T P.863.2 describes a set of models for predicting perceptual dimensions of degradations linked to the overall speech quality from narrowband (300 to 3 400 Hz) to fullband (20 to 20 000 Hz) telecommunication scenarios. The predictions target user judgements on four perceptual dimensions, as obtained in a subjective test described in an annex. The models described in Recommendation ITU-T P.863.2 are partially based on internal parameters of the model given in Recommendation ITU-T P.863. Recommendation ITU-T P.863.2 presents a detailed description of all model parts that are not contained in Recommendation ITU-T P.863. A conformity testing procedure is also specified in an annex to allow a user to validate whether an alternative implementation of the models is correct.