AI_LectureNote: A Retrospective Pilot Study of a Post-ASR Workflow for English-Script Rendering and Semantic Drift in Korean-English Medical Lectures
Computation and Language
Summary
The authors developed AI_LectureNote, a method to clean up speech-to-text results from Korean-English medical lectures by fixing how medical terms are shown. They tested it on four lectures and found it improved the display of English medical terms but sometimes changed the meaning or sentiment of sentences. Their study highlights that making the text look accurate doesn’t always keep the original meaning, and they recommend checking different types of errors separately. This work is a small pilot focused on finding specific problems rather than measuring overall error rates.
Authors
Kyeongeon Lee, Donghoon Chang, Seungryeol Baek, Taehong Kim, Wonjun Yang
Abstract
AI_LectureNote is a historical, readability-oriented post-ASR workflow for Korean-English medical lectures. It rewrites speech-to-text output into study transcripts while restoring Latin-script medical terms rather than Korean phonetic transliterations. We retrospectively evaluate the workflow on four author-recorded lectures across five conditions. In this pilot, post-processing raised the macro English-script rendering rate from 0.39 to 0.71 on the whisper-1 path and from 0.26 to 0.65 when applied to 3-minute chunked gpt-4o-transcribe output. However, English-script rendering did not imply semantic faithfulness: the two post-processed conditions showed semantic drift in 34 and 36 of 282 reference sentences and polarity failures in 11 and 13 of 101 polarity-cue rows. A descriptive cross-input comparison suggested different candidate failure patterns: polarity-failure sets overlapped more strongly across front-ends (Jaccard 0.60; 9 shared of 15 unioned failures) than general semantic-drift sets (Jaccard 0.23; 13 shared of 57 unioned drifts). This single-annotator pilot documents concrete failure modes rather than population rates and supports evaluating surface accuracy, term-script rendering, chunk-level script consistency, and medical-meaning preservation separately.