While perhaps not as much fun as changing piano sounds to speech, sometime ago Jim Bashford and I reported a slight of band that makes it possible to transform intelligible whispered discourse to an intelligible voiced monotone. This magic was accomplished by excising a 4 ms segment of the whispered speech and repeating it three times to produce four statements lasting 16 ms; then capturing another 4 ms of whisper which was repeated to produce another set of 4 statements, etc. The resulting voiced speech was quite intelligible despite the discard of 12 ms speech segments and the nonselective voicing of all phonemes. Presto change-o, it was possible to obtain a sex change by altering the duration of the repeated excised segments from 4 to 8 ms, and to avoid a monotone by slight variations in the durations of successive iterated captures.
Warren, R.M., and Bashford, J.A., Jr. 1978. Production of white tone from white noise and voiced speech from whisper. Bulletin of Psychonomic Society, 11, 327-329.
Richard M. Warren
and Distinguished Professor Emeritus
Department of Psychology
University of Wisconsin-Milwaukee
PO Box 413
Milwaukee, WI 53201