“…It is apparent, however, that emotion is conveyed by changes in pitch, timing and voice quality, which occur in different combinations for different emotions. Using current knowledge of emotions in human speech (Murray & Arnott, 1995), some prototype synthetic speech-with-emotion systems, including HAMLET, have been developed using existing synthesisers based on formant (Murray and Arnott, 1993, Cahn, 1990, Lewis & Tatham, 1995, Carlson, 1992 and concatenative technology (Murray and Edgington, 1998) . A review of progress and future possibilities in this area is given in (Murray et al, 1996).…”