My English sounds better than yours: Second-language learners perceive their own accent as better than that of their peers

Abstract

Second language (L2) learners are often aware of the typical pronunciation errors that speakers of their native language make, yet often persist in making these errors themselves. We hypothesised that L2 learners may perceive their own accent as closer to the target language than the accent of other learners, due to frequent exposure to their own productions. This was tested by recording 24 female native speakers of German producing 60 sentences. The same participants later rated these recordings for accentedness. Importantly, the recordings had been altered to sound male so that participants were unaware of their own productions in the to-be-rated samples. We found evidence supporting our hypothesis: participants rated their own altered voice, which they did not recognize as their own, as being closer to a native speaker than that of other learners. This finding suggests that objective feedback may be crucial in fostering L2 acquisition and reduce fossilization of erroneous patterns.

Do faces speak volumes? Social expectations in speech comprehension and evaluation across three age groups

“…In the repetition task, participants transcribed or repeated utterances (in three speech contexts) that were presented to them along with a static photograph of a given talker (six photographs). To control for inter-speaker variability in the three speech contexts, one speaker for each accent (Standard German, Korean-accented German, Palatinate regional accent of German) was used [for more details on the recordings, see 52 ], and two different versions of each recording were created by minor changes to the effective vocal-tract lengths (see also [ 19 , 53 ]). This allowed for a given experimental list to contain each accent paired with each ethnicity (e.g., the foreign-accented talker in voice version 1 was paired with an Asian prime, while the same foreign-accented talker in voice version 2 was paired with a white European prime).…”

“…Using Praat [ 55 ], the f0 and formant dispersion (but not duration or pitch) of the original recordings were changed to obtain two different versions of each voice. The recordings were altered by changing effective vocal-tract lengths (for a similar method, see [ 19 , 53 ]). To achieve this, the f0 of the original utterance was multiplied by a factor of 0.95 (to create voice version 1) and 1.05 (to create voice version 2).…”

Speak like a Native English Speaker or Be Judged: A Scoping Review

“…Despite efforts at multicultural education in the field of accents, a troubling level of non-native accent prejudice persists among speech language pathology students and practitioners [ 2 ]. Although these people have the privilege to be inducted into multicultural education, they remain prejudiced toward non-native accent speakers.…”

“…A listener’s view of the speaker’s accent can affect how he or she sees the person [ 1 ], which creates the first impression of an individual. The term accent has several meanings, but in speech it is an identifiable style of pronunciation that varies geographically or socioeconomically [ 1 ] and is one of the most noticeable features of speech [ 2 ]. Regional accents are influenced by geographical location, whereas social accents reflect speakers’ educational, socioeconomic, and ethnic backgrounds [ 2 , 3 ] as well as the phonetic variances caused by first language (L1) effect on the second language (L2) [ 2 ].…”

Effects of benchmarking and peer‐assessment on French learners' self‐assessments of accentedness, comprehensibility, and fluency

“…To date, only a handful of studies have focused on learners' self‐assessment of global dimensions of L2 speech, and such studies have revealed discrepancies between self‐ and other‐assessments. For example, Mitterer et al (2020) investigated L2 speakers' bias in rating their own versus fellow L2 speakers' accentedness. In this study, 24 female learners of L2 English consistently rated their own productions, which were acoustically altered to sound like male speakers, as being less accented than the productions of other (male‐sounding) learners, without realizing that the preferred voice actually belonged to them.…”