Judith C. Brown
Media Lab., MIT, Cambridge, MA 02139 Wellesley College, Wellesley, MA 02181
MIT, Cambridge, MA 02139
The determination of the pitch center of sounds that are frequency modulated has been the focus of a number of previous studies. The sources have usually been pure tones or synthetic complex sounds with a well-defined spectral composition. In a musical context these synthetic sounds differ in temporal and spectral properties from the natural sounds produced by stringed instruments; and it is these natural sounds which performers are trained to produce and to perceive in order to make intonation choices. Samples consisting of approximately 1 s of acoustic sounds produced by a virtuoso violist playing the notes D4, C5#, A5, and G6 with and without vibrato have been chosen for this study. The sounds without vibrato were then resampled to give frequencies from -15 to +21 cents, with respect to the geometric center of the sound, with vibrato. Experiments with forced choice for pairs of sounds using musically experienced listeners (including the violist who produced the original sounds) as subjects will be reported. These experiments include a control set consisting of the comparison of pitch levels of these same sounds played without vibrato. Preliminary results with listeners who are not string players are in accord with previous results, finding that the pitch perceived is that of the geometric average.