ABSTRACT
The three papers I have been asked to review are all concerned with the operations performed on the speech signal to extract information about its phonetic content, that is, the string of phonetic segments contained in the sound. In all three papers it is assumed that some transformation of the signal needs to be made initially in order to provide a representation or a display from which appropriate attributes or cues can be extracted. The three approaches each use a different display, and therefore differ in terms of the attributes that are extracted to derive a sequence of phonetic segments, but each approach contributes some insights into how these operations might be performed in the auditory and speech-perception system of the human listener.
