In earlier reports we have concluded that contextual invariance of categorical and magnitude estimates of speech quality could be improved by introducing a reference system and by normalizing the results with respect to it. In this study we investigate the possibility of substituting an actual reference signal with an "internal" reference. It is also studied whether in magnitude estimations a cross-modality matching using lines on a computer screen could be employed. Keywords: Speech quality, Speech synthesis, Magnitude estimations, Categorical estimations.