Three subjects have undergone the Tomatis® audio‐vocal training, which is a method that filters sidetone for auditory feedback. Long‐term average spectra of voiced and whispered speech was taken before and after the training with the following results: (1) Decrease in energy in the 50‐ to 200‐Hz zone; (2) increase in energy in the 800‐ to 1200‐Hz zone; (3) different individual overall energy shifts; (4) larger and wider formantic peaks in the long‐term average spectra; (5) increase in global energy; (6) greater pitch and power variability; and (7) increase in average pitch. A discussion concerning the employment of auditory feedback techniques in voice training follows. Future research should determine which psychophysiological characteristics of individuals determine their reactions to auditory feedback. It is also necessary to establish acoustic standards that could indicate desirable directions for voice training.
