|
Title:
|
STUDY ON AUTOMATIC GENERATION OF LECTURE VIDEOS BASED ON CONTENT ANALYSIS OF LECTURE SLIDES |
|
Author(s):
|
Kota Mizukami, Daisuke Deguchi, Tomoki Toda, Hiroshi Murase, Haruya Kyutoku and Tsubasa Minematsu |
|
ISBN:
|
978-989-8704-72-6 |
|
Editors:
|
Demetrios G. Sampson, Dirk Ifenthaler and Pedro Isaías |
|
Year:
|
2025 |
|
Edition:
|
Single |
|
Keywords:
|
Automatic Lecture Video Generation, Text Highlighting, Large Language Model |
|
Type:
|
Short Paper |
|
First Page:
|
353 |
|
Last Page:
|
356 |
|
Language:
|
English |
|
Cover:
|
|
|
Full Contents:
|
if you are a member please login 
|
|
Paper Abstract:
|
With the spread of online education, the demand for lecture videos in which instructors read aloud from slide decks has been increasing. However, creating such videos entails significant labor and cost. This study proposes a method that automatically generates lecture videosincluding narration audio and highlighted textusing the slides themselves as input. First, the slide content is analyzed with large language models and optical character-recognition. Based on the extracted information, narration speech is synthesized and key text within the slides is highlighted to enhance learners' understanding. Experiments using real presentation slides confirmed that the proposed method can automatically generate lecture videos. |
|
|
|
|
|
|
