Abstract

UC Berkeley Phonology Lab Annual Report (2007) Social judgments and their acoustic cues in read speech ∗ Molly Babel University of California, Berkeley Introduction In perceiving speech, listeners understand the talker’s intended message by extracting pho- netic segments and suprasegmental information from the acoustic signal and mapping the information onto meaningful chunks in the lexicon. The acoustic signal, however, contains talker-specific characteristics as well, including phonetic cues for typical sociolinguistic in- dexical variables such as gender, regional identity, and age. Listeners process this type of type of information, storing it in memory (Palmeri et al. 1993). To some extent, these talker-specific characteristics can be discussed as single isolated social categories, and gender may be the most well-studied of these characteristics. Speakers encode their gender in speech production from early childhood before any of the sexual dimorphic physiological changes take place that allow for men, on average, to be larger than women (Sachs et al. 1973, Perry et al. 2001). This gender information is used by listeners in processing speech sounds. Strand and Johnson (1996) synthesized the American English fricatives /s/ and /S/ on a seven point continuum and appended the sounds to naturally produced vowel-consonant closures from both male and female talkers that ranged from sod to shod. Male talkers elicited more sod responses than females, presumably due to listeners’ expectations that men, with typically larger vocal tracts, should produce fricatives with lower spectral center of energy than women (p. 18). In a second experiment, Strand Thanks very much to Ron Smyth and Henry Rogers for providing me with their database of recordings. I would also like to extend my gratitude to Keith Johnson and John Ohala for providing me with insightful comments that improved this paper immeasurably. This work could not have been done without research participants; thank you all for volunteering your time.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call