Multi-modal machine learning for business-critical insights in video conversations

The team is building a machine learning platform and solution to extract meeting insights from online meetings. Meeting insights denote moments from these meetings that may impact the company’s future product features, revenue, and customer satisfaction. This platform is driven by the market created by the widespread adoption of online virtual meetings as the main means of reaching clients in recent years. The internship will focus on developing a multi-modal solution that takes audio, visual, and its transcriptions as features and outputs moments of key insights. Specifically, the detection task is to detect interruptions, which identifies video and audio snippets where individuals are talking over another happening in business meetings.

Faculty Supervisor:

Scott Sanner;Anthony Bonner

Student:

Partner:

Talka AI Canada

Discipline:

Computer science

Sector:

Professional, scientific and technical services

University:

University of Toronto

Program: