This collaborative project improves access to information in videos by adding indexing, captioning, and search capability. The project goal is to automatically generate enhanced lecture videos, with logical index points, text captions, and keyword search capability, and evaluate their usefulness in light of students' learning outcomes and life situations in undergraduate STEM coursework across multiple departments and institutions. Deployment and evaluation of these Indexed, Captioned, and Searchable Tablet PC videos (ICS videos) spans more than 30 distinct courses in Computer Science, Biology & Biochemistry, and Geosciences at multiple campuses. In addition, automated captioning can dramatically increase the usefulness of main-stream videos for deaf and hearing-impaired students. The accessibility of ICS videos is investigated with a local school for deaf students.
Intellectual Merit: Key features of these ICS videos are: 1) Index points: Recorded video lectures have a list of logical index points, each in the form of a snapshot representing a video segment; 2) Text captions: The entire audio stream, including the lecture and student interaction, is presented as text in a separate panel; 3) Keyword search: Video segments that contain search keywords in the audio are displayed, sorted by keyword frequency; and 4) Evaluation: A rigorous evaluation is conducted in a range of disciplines with thousands of students.
Broader Impacts: The software is publicly available and support is provided to any interested educational institution. The project proposes a new and powerful model for online delivery of STEM coursework. ICS videos have the potential to make even more dramatic changes to distance education and education for the deaf by making mainstream lecture material accessible.
