Problem Statement
Multilingual AI Voice Chatbot
A voice assistant that listens to user queries in multiple languages. Understands the query and maintains a predefined personality. Generates text-based replies and converts them to speech in the same language. Handles at least a few languages and responds consistently according to its character.
Video Mood & Stress Detection
A system that captures real-time video of a person. Detects and tracks faces, extracts facial landmarks and micro-expressions. Classifies emotional states (e.g., happy, sad, anxious, stressed). Displays a real-time mood or stress score/dashboard.
Focus & Device Detection for Online Exams
A system that monitors a student via webcam during an exam. Tracks head pose and gaze to estimate focus (focused, distracted, away). Detects mobile phones or unauthorized devices in the frame. Computes a live attention/focus score and displays alerts on a dashboard.
Voice-Based Authentication & Spoof Detection
A system that authenticates users based on their voice. Detects spoofed or recorded voice attempts to prevent fraudulent access. Extracts audio features (MFCC, pitch, spectral features) for analysis. Provides real-time access control or alerts, optionally showing confidence scores on a dashboard.