EDM
Improving Speech Recognition of Named Entities in Classroom Speech with LLM Revision and Phonetic-Semantic Context
Wed Jul 1, 11:25 AM–11:40 AM · North 210
Part of Multimodal Analytics
Multimodal Learning Analytics Generative AI & Large Language Models NLP & Discourse Analysis Classroom Orchestration
★ Notable speakers
Jacob Whitehill
★
— Automated facial expression analysis; student engagement detection from video; affective computing for education
Improves classroom speech recognition of named entities — student and subject names — by combining LLM-based revision with phonetic-semantic context signals.