Second International Conference on Spoken Language Processing (ICSLP'92)
Banff, Alberta, Canada
Current synthetic speech systems produce voice output which is unacceptable for general public use. Listeners report it sounds mechanical, machine-like and is not pleasant to listen to. A model is being developed which adds variability to the wave-form; this variability conveys emotion or attitude and produces more natural sounding synthetic speech.
Bibliographic reference. Morton, Katherine (1992): "Adding emotion to synthetic speech dialogue systems", In ICSLP-1992, 675-678.