5th International Conference on Spoken Language Processing

Sydney, Australia
November 30 - December 4, 1998

Acoustic-Articulatory Evaluation of the Upper Vowel-Formant Region and its Presumed Speaker-Specific Potency

Frantz Clermont (1), Parham Mokhtari (2)

(1) University of Tsukuba, Japan
(2) LORIA-Campus Scientifique, France

We present some evidence indicating that phonetic distinctiveness and speaker individuality, are indeed manifested in vowels' vocal-tract (VT) sha pes estimated from the lower and upper formants, respectively. The methodology developed to demonstrate this dichotomy, implicates Schroeder's (1967) acoustic-articulatory model which can be coerced to yield area-function approximations to VT-shapes of differing formant components. Using ten steady-state vowels recorded in /hVd/-context, five times at random, by four adult-male speakers of Australian English, VT-shape variability was then measured on an intra- and an inter-speaker basis. Gross shapes estimated from the lower formants, caused the largest spread amongst the vowels of individual speakers. By contrast, more detailed sha pes estimated from certain higher formants of front and back vowels, caused the largest spread amongst speakers. These results contribute a quasi-articulatory substantiation of a long-standing view on the speaker-specific potency of the upper vowel-formant region, together with some useful implications for speech and speaker recognition.

Full Paper

Bibliographic reference.  Clermont, Frantz / Mokhtari, Parham (1998): "Acoustic-articulatory evaluation of the upper vowel-formant region and its presumed speaker-specific potency", In ICSLP-1998, paper 0087.