Multimodal Context Extraction for Virtual Meetings

During the pandemic situation the use of online meetings has increased drastically and in the education and corporate industry it has been a key way to communicate and interact. Huge amount of unprocessed information is present in the videos which can be of great help. Maintaining a report or summary of such virtual meetings manually is tedious and inefficient although many platforms allow recording such meetings but it becomes inefficient to maintain and understand those recordings. A well formatted report summary which can be auto-generated from the recordings and can be easily shared could solve the above issues. To improve the virtual teaching and learning experience system would be generating multiple choice questions based on the extracted keyword from the generated summary.