Psychometric Functions of Vowel Detection and Identification in Long-Term Speech-Shaped Noise

Abstract

Purpose The goal of this study was to investigate vowel detection and identification in noise and provide baseline data regarding how vowel perception changed with signal-to-noise ratios. Psychometric functions of vowel detection and identification for 12 American English isolated vowels in long-term speech-shaped noise were examined for young listeners with normal hearing in this study. Method Vowel detection was measured at sensation levels from −10 to +5 dB (re: thresholds of vowel detection from the study of Liu and Eddins, 2008a ) with a 4-interval forced-choice procedure. Thresholds of vowel detection were computed for each listener as the speech level at which 70.7% correct performance was reached. Vowel identification was then examined at sensation levels from 0 to 12 dB relative to detection thresholds for each listener. Thresholds of vowel identification were calculated as the speech level with vowel identifiability ( d ′) equals to 1. Results Thresholds of vowel detection and identification were significantly affected by vowel category. Slopes of psychometric functions of vowel identification were significantly dependent on vowel category, whereas slopes of psychometric functions of vowel detection were not. Conclusions These results suggest that, given the same sensation levels, especially at low sensation levels, vowel sounds are not equally perceivable in terms of identifiability.

The perception of artificial-intelligence (AI) based synthesized speech in younger and older adults

“…Presentation of a low number of masking levels can limit interpretations about whether AI speech is less intelligible than human speech, because intelligibility differences can be related to differences in speech-perception thresholds or speech-perception variability (i.e., the steepness of the relation between masking levels and intelligibility). Measuring the whole psychometric intelligibility continuum from very low to very high speech intelligibility (Wingfield et al, 2000;MacPherson and Akeroyd, 2014;Liu and Jin, 2019;Masalski et al, 2021;Ross et al, 2021;Irsik et al, 2022) enables a more detailed characterization of potential differences between AI and human speech. Finally, overall differences between computer-generated speech and human speech are also hard to interpret because speech materials typically vary on a number of acoustic parameters that may be unspecific to the human-vs-AI-speech contrast but could also be present for different human voices.…”