Vocal fry in realistic speech: Acoustic characteristics and perceptions of vocal fry in spontaneously produced and read speech

Abstract

There has been a great deal of recent interest in vocal fry, both in production and perception. However, much of the scientific literature that has used naturally produced fry has focused on speech elicited through reading, rather than spontaneous speech. The current study compares reading with spontaneous speech elicited in various ways for both male and female speakers, recorded in dyads. The speakers were asked to teach their partner information, give instructions, and describe their qualifications for a scholarship, with reading controls for each task. Surprisingly, we found more vocal fry (in both proportion of words and words/duration) for men than women. Men differed in the amount of fry across task types. Interactions of speaker sex, task type, and fry were examined for measures of F0, pitch range, intensity level, jitter, shimmer, and HNR. Listener perceptions of speech with and without vocal fry were also obtained. Task type affected listener perceptions of speech with vocal fry. Vocal fry in spontaneously produced speech seems to differ from fry in speech produced during reading, in quantity, acoustics, and listener perceptions. Additionally, acoustic measures suggest that vocal fry may be intimately tied to decreased vocal effort across task types and speaker sex.

The Effects of Pitch, Gender, and Prosodic Context on the Identification of Creaky Voice

“…Though the female speakers have significantly more creak, the difference is still quite small. In contrast, Irons and Alexander (2016) analyze 16 American speakers between 18 and 22 years producing Davidson both read and spontaneous speech, and find that about 25% of all speech produced by the male speakers contained creak whereas the comparable amount for women was about 9%. Using an automated detection algorithm, Abdelli-Beruh et al (2016) find that there is no difference between male and female American English speakers reading the Grandfather Passage.…”

The versatility of creaky phonation: Segmental, prosodic, and sociolinguistic uses in the world's languages

“…Despite such comments, a review of research on the distribution of creaky phonation in the speech of male and female American speakers does not find good evidence of an overwhelming prevalence of creaky phonation among (young) female speakers. Some studies suggest that overall women produce more syllables in read sentences or passages with creak than men do (Abdelli‐Beruh et al, 2014; Wolk et al, 2012; Yuasa, 2010) (though these studies mainly use a perceptual rather than acoustic criteria to identify creak), while others demonstrate that differences between men and women are either small or absent altogether (Abdelli‐Beruh et al, 2016; Irons & Alexander, 2016; Melvin & Clopper, 2015; Pratt, 2018). In fact, the Southern United States male speakers in Irons and Alexander (2016) actually produced proportionally more syllables with creaky phonation (26%) than the female speakers did (8%).…”

“…Some studies suggest that overall women produce more syllables in read sentences or passages with creak than men do (Abdelli‐Beruh et al, 2014; Wolk et al, 2012; Yuasa, 2010) (though these studies mainly use a perceptual rather than acoustic criteria to identify creak), while others demonstrate that differences between men and women are either small or absent altogether (Abdelli‐Beruh et al, 2016; Irons & Alexander, 2016; Melvin & Clopper, 2015; Pratt, 2018). In fact, the Southern United States male speakers in Irons and Alexander (2016) actually produced proportionally more syllables with creaky phonation (26%) than the female speakers did (8%). As for spontaneous speech, a study using sociolinguistic interviews found greater proportions of creaky phonation for female (27%) than male (8%) American speakers, regardless of whether they were White or Black Washington, D.C. speakers (Podesva, 2013).…”