Modeling the role of inherent spectral change in vowel identification

Abstract

Statistical analysis of F1 and F2 measurements from nucleus and offglide sections of isolated Canadian English vowels shows significant formant frequency change not only for the ‘‘phonetic diphthongs’’ /e/ and /o/, but also for the ‘‘monophthongs’’ /ι/, /q/, and /1/. In a perceptual experiment, brief sections were extracted from ‘‘nucleus’’ and ‘‘offglide’’ portions of naturally produced vowels. Two sections from each vowel were presented to listeners in each of three conditions: (1) natural order (nucleus followed by offglide); (2) repeated nucleus (nucleus followed by itself); and (3) reverse (offglide followed by nucleus). Listeners’ error rates for the natural order condition were comparable to those for unmodified full vowels (averaging 14% and 13%, respectively). Significantly higher error rates were found for the repeated nucleus (32%) and reverse (38%) conditions. Observed confusion matrices were strongly correlated with predictions from a pattern recognition model incorporating the formant measurements. This study provides evidence for the importance of inherent spectral change in listeners’ perception of isolated vowels. In addition, the problem of the parametric representation of formant trajectories is discussed and preliminary evidence for the persistence of vowel-inherent spectral change in consonantal context is presented.

Analysis and analogy in the perception of vowels

“…Independently of pitch matching studies, though, the premise that vowel contrasts are equivalent to simple contrasts of stationary pitch is not well supported (see Nearey & Assmann, 1986), nor have campaigns to describe multimodal speech perception settled on an auditory pitch space as the common metric (see Summerfield, 1987;Watson, Qiu, Chamberlain, & Li, 1996). In this light, the present results offer a way to explain the ability of subjects to form pitch analogies to vowels without invoking a perceptual pitch representation as an intermediate form.…”

“…This is a ranking of vowels considered in the ideal. In actual practice, spoken vowels are far from stationary in their spectral manifestations (Nearey & Assmann, 1986) and extensive overlap is observed in the distributionsof second formant frequency values of different vowels.…”

Temporal cues for vowels and universals of vowel inventories

“…But is was not only contrasts such as /ey eh/, where one was a clear diphthong, which improved with a later gate point; contrasts increased across the board. This reinforces the discovery by Nearey and Assmann [7] that most North American English vowels have some degree of diphthongization which, though slight in some cases, is perceptually important. …”

Perception in a Variable but Structured World: The Case of Speech Perception

“…Additionally, the realization of phonetic categories depends on the linguistic context that those sounds are produced in (Liberman et al, 1967), and a variety of models have been proposed to account for how categorization of one segment is affected by adjacent segments (Massaro, 1987;Nearey, 1997;Nearey & Assman, 1986;Oden & Massaro, 1978;Smits, 2001;Sonderegger & Yu, 2010). …”