This dataset derives from the Master’s dissertation of Sara Gomes Rosa and contains acoustic and auditory-perceptual measures extracted from speech samples recorded through two transmission channels (direct recording and telephone transmission). The dataset includes two processed versions of the acoustic measures generated from the same speech corpus using different extraction parameter configurations. Variables comprise measures of fundamental frequency, frequency variability, voice quality, and spectral parameters, in addition to structured auditory-perceptual evaluations. All data have been anonymized.