The phonological function of vowels is maintained at fundamental frequencies up to 880 Hz

Abstract

In a between-subject perception task, listeners either identified full words or vowels isolated from these words at F0s between 220 and 880 Hz. They received two written words as response options (minimal pair with the stimulus vowel in contrastive position). Listeners' sensitivity (A') was extremely high in both conditions at all F0s, showing that the phonological function of vowels can also be maintained at high F0s. This indicates that vowel sounds may carry strong acoustic cues departing from common formant frequencies at high F0s and that listeners do not rely on consonantal context phenomena for their identification performance.

Vowel recognition at fundamental frequencies up to 1 kHz reveals point vowels as acoustic landmarks

“…It seems, therefore, likely that listeners used the vowel /o/ as a substitute because /u/ was not presented to them as a response option. The results by Friedrichs et al (2015a), who found the same eight vowels used in the present study identifiable up to an f o of 880 Hz when recorded in minimal pairs and tested in a two-alternative forced choice task, could also be explained within this context. As a single talker was asked to produce several different twoword combinations containing a vowel in contrastive position (e.g., the German words Buden vs Boden), it is possible that the talker produced vowels with acoustic features alike or different from those of a point vowel at higher f o s to make them distinguishable (e.g., producing an /o/ more toward /a/ to distinguish it from /u/).…”

“…In a more recent study, Maurer et al (2014) investigated the high-pitched vowels /i y oe a O u/ produced by a female Cantonese opera singer in isolation and monosyllabic consonant-vowel utterances and found that /i a O u/ could be identified by more than 80% of the listeners within an f o range of 820-860 Hz. In a study using a two-alternative forced choice task, Friedrichs et al (2015a) provided evidence that the phonological function of the eight vowels /i y e ø e a o u/ (i.e., the function they fulfil in linguistic contrastive position to help listeners distinguish between words) can be maintained at f o su pt oa tl e a s t8 8 0H zw h e n they were produced in minimal pairs. These judgments were made on excised steady-state vowel nuclei (250 ms) excluding consonantal context phenomena such as co-articulation and formant transitions.…”

Perceptual clustering of high-pitched vowels in Chinese Yue Opera

“…Previous studies have indicated that different tasks do affect the listener's identification performance. For instance, participants performed better when they were presented with more meaningful response options (e.g., written words containing the target vowel vs. vowel letters), fewer response options, and a lower degree of talker variability (Friedrichs et al, 2017(Friedrichs et al, , 2015a(Friedrichs et al, , 2015b.…”

“…For example, Friedrichs et al (2015a) found that the phonological function of the isolated steady-state vowels /i y e ø ɛ a u o/ can be maintained at fos up to 880 Hz when they were tested in a listening test with only two response options. In a follow-up study investigating the influence of talker variability, Friedrichs et al (2017) found that the cardinal vowels (point vowels) /i a u/ remained identifiable even up to 1046 Hz when they were tested in isolation and multiple response options were provided.…”

Influences of Fundamental Oscillation on Speaker Identification in Vocalic Utterances by Humans and Computers

“…However, the finding might might be in line with recent observations that vocal tract detail in vocalic utterances has been underestimated in vowels produced at high fo . [6] showed that vowels produced by soprano voices still contain sufficient information about vowel identity, even though formant information is vastly deteriorated. It will be interesting to study whether speaker specific information is also maintained at higher frequencies in human listener.…”