UTILIZATION OF ENGLISH LANGUAGE SPEECH ASSISTANT FOR PUBLIC SPEAKING SKILLS OF FIRST-YEAR TOURISM STUDENTS IN BATANGAS STATE UNIVERSITY, MALVAR CAMPUS
This study examined the effectiveness of ELSA in enhancing public speaking skills among first-year Tourism students at Batangas State University–Malvar Campus. Using a quasi-experimental design, two groups were compared: a control group that received traditional instruction and an experimental group that utilized ELSA, an AI-driven speech tool that provided real-time pronunciation feedback. Both groups underwent pre-tests and post-tests, assessed using validated rubrics measuring fluency, pronunciation, vocabulary, grammatical accuracy, body language, spatial awareness, eye contact, clarity, organization, relevance, confidence, audience engagement, and message delivery. Findings revealed that the experimental group demonstrated significant improvements. For fluency, the experimental mean was 2.99, exceeding the control group's 2.51 (t = –4.675, p < 0.000). Pronunciation improved, with the experimental mean at 3.11, surpassing the control's 2.40 (t = –4.675, p < 0.000). Vocabulary scores rose from 2.82 (control) to 3.25 (experimental) (t = –3.920, p < 0.001). Grammatical accuracy increased from 2.79 to 3.21 (t = –3.452, p < 0.002). Message delivery improved to 2.89 compared to 2.31 (t = –2.324, p < 0.024). These results confirmed ELSA's effectiveness in enhancing speech clarity, confidence, and audience engagement.The study underscored the role of AI-driven speech tools in equipping students with essential communication skills for global tourism careers. Based on the findings, ELEVATE (Empowering Learners’ English Voice through AI-Technology and Engagement) was proposed to support educators in integrating AI-assisted speech training into higher education.
Publication Details
- Journal
- Zenodo (CERN European Organization for Nuclear Research)
- Published
- 2025-06-22
- DOI
- https://doi.org/10.5281/zenodo.15714291
- Primary Topic
- AI in Service Interactions
- Type
- article
- Field-Weighted Citation Impact
- 0.00