FROM FAMILY INTERACTION TO DYNAMIC FEEDBACK:ADVANCES IN MANDARIN LANGUAGE LEARNING APPLICATIONS

Mandarin Chinese is vital as Singapore's second most spoken language, but pronunciation training remains underrepresented in learning. This thesis covers two Computer-Assisted Language Learning (CALL) projects enhancing Mandarin acquisition. The first focuses on a tablet app for children and parents, using Automatic Speech Evaluation (ASE) to give phoneme and tone feedback, encouraging parent-child interaction. The second project integrates a Large Language Model (LLM) into a language app for young adults, providing personalized feedback based on the Zone of Proximal Development (ZPD) theory. Both projects use mixed methods to evaluate effectiveness and user interaction. The findings aim to contribute to the fields of Human-Computer Interaction (HCI) and language education, offering insights into the co-use of technology for language learning in family settings and the potential of LLMs to support dynamic feedback in pronunciation training.

Author dc:creator

• SHI MINGQIAN

• Large Language Model
• Computer-Assisted Pronunciation Training
• Computer-Assisted Language Learning
• Human-Computer Interaction

Licence

• https://scholarbank.nus.edu.sg/bitstreams/f90c9d46-505d-4c2b-ba71-d267fb0b0e13/download