Clear speech in Tashlhiyt Berber: The perception of typologically uncommon word-initial contrasts by native and naive listeners

Abstract

Tashlhiyt Berber is known for having typologically unusual word-initial phonological contrasts, specifically, word-initial singleton-geminate minimal pairs (e.g., sin vs ssin) and sequences of consonants that violate the sonority sequencing principle (e.g., non-rising sonority sequences: fsin). The current study investigates the role of a listener-oriented speaking style on the perceptual enhancement of these rarer phonological contrasts. It examines the perception of word-initial singleton, geminate, and complex onsets in Tashlhiyt Berber across clear and casual speaking styles by native and naive listeners. While clear speech boosts the discriminability of pairs containing singleton-initial words for both listener groups, only native listeners performed better in discriminating between initial singleton-geminate contrasts in clear speech. Clear speech did not improve perception for lexical contrasts containing a non-rising-sonority consonant cluster for either listener group. These results are discussed in terms of how clear speech can inform phonological typology and the role of phonetic enhancement in language-universal vs language-specific speech perception.

Linguistic disparities in cross-language automatic speech recognition transfer from Arabic to Tashlhiyt

“…Only in Clear speech, where ASR performance was enhanced, do we observe that vowelless words are accurately transcribed at lower rates than voweled words. This interaction parallels recent findings examining perceptual patterns of clearly- and casually-produced Tashlhiyt words by non-native listeners in 46 . In that study, Tashlhiyt-naive listeners showed a clear speech boost in perceiving Tashlhiyt words that have phonotactic patterns that are also present in English (i.e., clear speech enhanced the discrimination of pairs like sin vs. fin ).…”

“…Thus, we also compare cross-language ASR transfer across clear and reduced speech productions of the target language. It is predicted that the ASR system will perform less well for reduced speech forms, consistent with prior findings that fast, casual speech is less well understood than clear speech by human (e.g., for Tashlhiyt 46 ) and machine 42 comprehenders. However, if vowelless words are understood at an even lower rate than voweled words in one speech mode, this can further reveal how disparities across phonotactic patterns in the target language can be amplified given the range of variation in speaking styles that is found across users and contexts.…”