Automatic estimation of the first three subglottal resonances from adultsa speech signals with application to speaker height estimation

Recent research has demonstrated the usefulness of subglottal resonances (SGRs) in speaker normalization. However, existing algorithms for estimating SGRs from speech signals have limited applicabilityathey are effective with isolated vowels only. This paper proposes a novel algorithm for estimating the first three SGRs (Sg 1 , Sg 2 and Sg 3) from continuous adultsa speech. While Sg 1 and Sg 2 are estimated based on the phonological distinction they provide between vowel categories, Sg 3 is estimated based on its correlation with Sg 2 . The RMS estimation errors (approximately 30, 60 and 100 Hz for Sg 1 , Sg 2 and Sg 3 , respectively) are not only comparable to the standard deviations in the measurements, but also are independent of vowel content and language (English and Spanish). Since SGRs correlate with speaker height while remaining roughly constant for a given speaker (unlike vocal tract parameters), the proposed algorithm is applied to the task of height estimation using speech signals. The proposed height estimation method matches state-of-the-art algorithms in performance (mean absolute error = 5.3 cm), but uses much less training data and a much smaller feature set. Our results, with additional analysis of physiological data, suggest the existence of a limit to the accuracy of speech-based height estimation.