Abstract
The present study refined Kondo and Yamamoto’s (2024) PPP-based pronunciation training program, implemented in a flipped-learning format. The revised version was applied to two groups of Japanese university students (with higher and lower proficiency) over one semester, and its effects were examined through two questionnaires and two speaking tests (reading aloud and picture description). A new pronunciation rating rubric was also developed to enhance inter-rater reliability. The findings indicated that the main revisions—teaching prosody before segmental phonemes, shortening the pronunciation practice component, and incorporating a pronunciation-rating app for self-monitoring—were received positively by the participants. Both proficiency groups significantly improved their speech comprehensibility and generally perceived the training as enjoyable and satisfactory. Moreover, the revised rubric yielded acceptable inter-rater reliability.
