N.V.J.K KartikKartik

Research system · IIT Bombay · 2023 — 24 · Audio · video · text

AffectBots

A multimodal tutoring system that reads audio, video and text to recognise a student’s emotion while a lesson is happening.

Plate: audio, video and text signals over one lesson.

What I did

I built AffectBots as part of the research group’s classroom AI work.

Who I made it with

Educational Technology group, IIT Bombay.

What it’s for

Built during my research role in IIT Bombay’s Educational Technology group, alongside the classroom diarization and shared-regulation work.

Why it was hard

Three signals that arrive at different rates and disagree with each other, fused fast enough to matter during the lesson rather than after it.

Receipts

  1. Previous portfolio, as it stood before this redesign (self-reported)Self-reported: IIIT Dharwad (Data Science & AI), research at IIT Bombay Educational Technology, CRF approach for SSMR.

Every source on the site

Next, a little further back in timeConversation analytics →