Charles Explorer logo
🇨🇿

Perceptual and acoustic correlates of Russian accent in Czech

Publikace na Filozofická fakulta |
2022

Tento text není v aktuálním jazyce dostupný. Zobrazuje se verze "en".Abstrakt

Sounding native in L2 is, for many learners, an almost unachievable goal, and even L2 users whose speech is highly intelligible can speak with a considerable degree of foreign accent [1]. Accent rating is affected by various phonemic and phonetic divergences [1-2]. Here we focus on L1 Russian learners of L2 Czech and aim to identify and quantify the cues that lead to perception of foreign accentedness. To assess the noticeability of deviations in their L2 speech, Russian learners of Czech (n=10, 5 F) were recorded reading a text in Czech. In a subsequent perception test, native Czech listeners (n=44) rated the extent of foreign accent on a continuum from 0 (no accent) to 100 (very strong accent) and commented which speech phenomena revealed the non-Czech origin of the speakers.

The most frequently noticed deviations pertained to word stress (its placement and acoustic realization) and, relatedly, to the realization of vowel quantity contrasts. In order to objectively quantify the degree of accentedness in word stress and vowel quantity, a contrastive acoustic analysis compared the 10 Russian speakers to the recordings of identical materials recorded by 10 native Czech speakers. The analysis revealed that the Russian speakers' relative vowel duration fell outside the native speakers' interquartile range. The placement and realization of word accent was influenced by L1 phonotactics and L2 orthography. The ratings of accentedness that were normalized by z-score strongly correlated with the total number of deviations that the native judges commented on (ρ=0.88). Word stress was previously reported as a predictor of high accentedness rating for L1 English speakers of L2 German [3], and for L2 English speakers with various L1s [4]. The contributions to foreign accent identified here, both segmental and suprasegmental, will be discussed and compared to findings from other languages.