Dinçer S., Mavaşoğlu M., Kurt Ç.
SAGE OPEN, cilt.16, sa.3, ss.1-17, 2026 (SSCI, Scopus)
-
Yayın Türü:
Makale / Tam Makale
-
Cilt numarası:
16
Sayı:
3
-
Basım Tarihi:
2026
-
Doi Numarası:
10.1177/21582440261492392
-
Dergi Adı:
SAGE OPEN
-
Derginin Tarandığı İndeksler:
Academic Search Ultimate (EBSCO), Social Science Premium Collection (ProQuest), Education Source Ultimate (EBSCO), Scopus, Social Sciences Citation Index (SSCI), Education Abstracts, ERIC (Education Resources Information Center), EBSCO Education Source, Directory of Open Access Journals
-
Sayfa Sayıları:
ss.1-17
-
Çukurova Üniversitesi Adresli:
Evet
Özet
This study explored the acoustic fluency and speaking anxiety patterns associated with an eight-week AI-mediated, task-based, individualized speaking practice package delivered through ChatGPT voice interaction. The package was implemented as an integrated whole among university-level learners of French as a foreign language. Grounded in Skill Acquisition Theory and Cognitive Load Theory, the study employed a pretest/posttest quasi-experimental design with intact classes (n = 61). The experimental group completed structured voice interactions with a fixed feedback sequence; the control group received duration- and task-matched instructor-led practice. Fluency was assessed through four Praat-derived acoustic indicators (speech rate, mean length of run, self-repair rate, and repetition frequency) and anxiety was measured via the Second Language Speaking Anxiety Scale. Effect sizes and Wild Cluster Bootstrap confidence intervals, complemented by ANCOVA models, revealed patterns favoring the experimental group for speech rate, mean length of run, and speaking anxiety, alongside a shift in the self-repair profile; repetition frequency yielded a weaker pattern. The asymmetric distribution across fluency subcomponents suggests that component-level acoustic measurement captures distinctions obscured by holistic ratings. Given the limited cluster structure (k = 4) and the integrated nature of the package, all findings should be considered exploratory and feasibility-oriented rather than confirmatory of any single component.