Perceiving the sex and identity of a talker without natural vocal timbre

Abstract

The personal attributes of a talker perceived via acoustic properties of speech are commonly considered to be an extralinguistic message of an utterance. Accordingly, accounts of the perception of talker attributes have emphasized a causal role of aspects of the fundamental frequency and coarsegrain acoustic spectra distinct from the detailed acoustic correlates of phonemes. In testing this view, in four experiments, we estimated the ability of listeners to ascertain the sex or the identity of 5 male and 5 female talkers from sinusoidal replicas of natural utterances, which lack fundamental frequency and natural vocal spectra. Given such radically reduced signals, listeners appeared to identify a talker's sex according to the central spectral tendencies of the sinusoidal constituents. Under acoustic conditions that prevented listeners from determining the sex of a talker, individual identification from sinewave signals was often successful. These results reveal that the perception of a talker's sex and identity are not contingent and that fine-grain aspects of a talker's phonetic production can elicit individual identification under conditions that block the perception of voice quality.What can a listener perceive in the speech of an unfamiliar talker? Even a brief utterance can convey a linguistic message and something about the talker who produced it. Although the perception ofpersonal attributes has commonly been explained by an account separate from the perception of linguistic properties, a recent study has shown that phonetic details can also be used to identify talkers and to distinguish them from one another (Remez, Fellowes, & Rubin, 1997). Surprisingly, when acoustic test materials forced performance to depend on phonetic attributes, listeners occasionally mistook male talkers for female talkers, and vice versa. The present report describes a series of experiments intended to clarify the interpretation ofthis counterintuitive finding, posing these questions: (1) Is the sex of a talker identifiable in a sine wave utterance replica? (2) Are differences across talkers in the central spectral tendency ofthe sinusoidal constituents responsible for differing impressions of the sex of a sine wave talker? (3) Are individuals identifiable under acoustic conditions that preclude the identification of sex?Many studies of talker recognition by ear, by automatic classification, or by visual inspection of spectroThis research was supported by Grants DC00308 (to R.E.R.) and HDOl994 (to Haskins Laboratories) from the National Institutes of Health.The authors gratefully acknowledge the meticulous assistance and trenchant advice of Chris Darwin, Steve Goldinger, Harry Levitt, Jennifer Lipton, Larry Rosenblum, Jim Sawusch, Dalia Shoretz, Saskia Smith, Steve Stroessner, Doug Whalen, and Fay Xing. Correspondence should be addressed to R. E. Remez, Department of Psychology, Barnard College, 3009 Broadway, New York, NY 10027-6598 (e-mail: remez@paradise.barnard.columbia.edu).grams have sought to tie variation across individu...

The role of f 0 and formant frequencies in distinguishing the voices of men and women

“…Schwartz and Rine (1968), for example, reported near-perfect judgments of talker sex from whispered vowels recorded from a small sample of men and women, and, using grade school-aged children as listeners, Bennett and Montero-Diaz (1982) reported reasonably accurate judgments of speaker sex for whispered vowels. Speaker sex can also be judged from isolated voiceless fricatives (Ingemann, 1968) and from sine wave replicas of short sentences (Fellowes, Remez, & Rubin, 1997). Sine wave replicas, being aperiodic, do not have pitch in the traditional sense.…”

Learning to recognize talkers from natural, sinewave, and reversed speech samples.

“…Is the perception of a talker's sex a prerequisite for individual identification? Fellowes et al (1997) addressed these questions using a second set of modified sex-neutral sinewave signals. Intertalker differences were completely eliminated from the tokens by rescaling the tone pattern of every talker to match the average formant values of all ten talkers.…”

“…The extent to which perception of talker identity is contingent on perception of talker sex was addressed directly by Fellowes et al (1997) using perceptual tests and different types of sinewave signals. The first set of sinewave tokens were the same sinusoidal replicas used in the current training experiments.…”

Neural Encoding of Speech and Music: Implications for Hearing Speech in Noise

“…The F 0 and lower harmonics, along with other acoustic ingredients, contribute to the perception of pitch. 59,60 Psychophysical experiments have demonstrated that F 0 differences facilitate the ability to distinguish between competing vowels or sentences. [61][62][63][64][65][66] The perceptual benefit provided by F 0 separation is apparent in neural processing.…”