Industry ProjectsJul'25 - Jun'26H-adGen: Framework for Hyperlocalized Personalization of Video AdvertisementsA multimodal AI framework that localizes pre-recorded video advertisements by replacing spoken brand and location-specific entities while preserving the original speaker’s voice and visual identity. Integrates speech recognition, zero-shot voice cloning, lip synchronization, facial restoration, and multilingual content moderation to generate regionally tailored advertisements in Hindi, Marathi, Tamil, and Telugu.Cere LabsResearch PaperAbout ProjectH-adGen: Framework for Hyperlocalized Personalization of Video AdvertisementsA multimodal AI framework that localizes pre-recorded video advertisements by replacing spoken brand and location-specific entities while preserving the original speaker’s voice and visual identity. Integrates speech recognition, zero-shot voice cloning, lip synchronization, facial restoration, and multilingual content moderation to generate regionally tailored advertisements in Hindi, Marathi, Tamil, and Telugu.ContributionsAddressed the scalability and localization challenges of celebrity-led advertising by developing a framework that adapts a single brand campaign for different retailers, locations, and promotional offers.Designed a brand-to-retailer personalization workflow enabling MSMEs to generate localized advertisement variants without requiring a separate video shoot for every variation.Architected a two-stage multimodal AI pipeline integrating 5+ components across speech recognition, voice generation, lip synchronization, facial restoration, and content moderation.Developed spoken-entity personalization capabilities to replace brand names, location references, and promotional offers while preserving the original speaker’s voice identity and synchronized lip movements.Enabled multilingual advertisement localization with support for 22 Indic languages, targeting regionally tailored advertising across diverse linguistic markets.Engineered an evidence-based benchmarking framework using 640 speakers across two language families to evaluate model suitability for multilingual speech applications.LinksResearch Paper
Personal ProjectsJul'25 - Jun'26AI-Enabled Real-Time Accident Monitoring & Response Coordination SystemAn AI-powered road accident detection and response coordination system that analyzes CCTV surveillance footage using YOLOv11 to identify vehicle collisions across 8+ accident categories. Integrates low-latency video processing, real-time incident alerts via WebSockets, a FastAPI backend, and a Next.js verification dashboard to streamline incident review and emergency response coordination.Personal ProjectView on GitHubAbout ProjectAI-Enabled Real-Time Accident Monitoring & Response Coordination SystemAn end-to-end accident monitoring system combining YOLOv11-based computer vision, live CCTV surveillance, real-time event communication, and an incident-verification interface to support faster emergency response.ContributionsDeveloped a real-time vehicle-collision detection system using YOLOv11 and CCTV surveillance footage to identify road accidents across 8+ accident categories.Optimized the detection workflow to achieve a reported processing cycle of 8 ms per frame, supporting low-latency video analysis.Built a FastAPI backend for accident-detection services and incident event handling.Implemented WebSocket-based real-time alerts to communicate detected incidents to the monitoring interface.Developed a Next.js and TypeScript frontend with a three-stage incident-verification workflow for reviewing and responding to detected accidents.Integrated authentication, PostgreSQL, and Prisma to support application access and persistent data management.Reported model performancemAP@5099%Reported object-detection performance across the evaluated accident classes.mAP@50–9589.3%Reported mean average precision across stricter IoU thresholds.Precision96.8%Reported proportion of predicted accident detections that were correct.Recall96.4%Reported proportion of ground-truth accident instances detected.LinksView on GitHub
