Acoustic differences between normally phonated and whispered speech

Whispered speech is different from normally phonated speech in more than just lack of vocal fold vibration. Previous research suggests that speakers also modify their speaking rate and formant frequencies. The extent of these differences may depend on speaker gender, but this has not been studied in American English. We examined differences in duration and acoustic vowel space between normally phonated and whispered speech. We recorded 17 ciswomen, 11 cismen, and 10 transwomen producing hVd words in both normally phonated and whispered speech. Formant frequencies for /ae/, /ɪ/, /ɑ/, and /u/ were bark-transformed and vowel space area was calculated. The results indicated differences in the formants across the two speaking conditions, both in the absolute shift of formant values and in the size of the vowel space. In addition, differences in speaking rate (measured as the duration of the /hV/ portion) also showed differences between the two speaking conditions across the three speaker groups.Whispered speech is different from normally phonated speech in more than just lack of vocal fold vibration. Previous research suggests that speakers also modify their speaking rate and formant frequencies. The extent of these differences may depend on speaker gender, but this has not been studied in American English. We examined differences in duration and acoustic vowel space between normally phonated and whispered speech. We recorded 17 ciswomen, 11 cismen, and 10 transwomen producing hVd words in both normally phonated and whispered speech. Formant frequencies for /ae/, /ɪ/, /ɑ/, and /u/ were bark-transformed and vowel space area was calculated. The results indicated differences in the formants across the two speaking conditions, both in the absolute shift of formant values and in the size of the vowel space. In addition, differences in speaking rate (measured as the duration of the /hV/ portion) also showed differences between the two speaking conditions across the three speaker groups.