Year-2025Year-2025http://repository.iiitd.edu.in/xmlui/handle/123456789/18062026-09-20T13:32:15Z2026-09-20T13:32:15ZProject ROBIFYJain, AryanGrover, Anuj (Advisor)http://repository.iiitd.edu.in/xmlui/handle/123456789/21682026-09-18T22:00:35Z2025-07-20T00:00:00ZProject ROBIFY
Jain, Aryan; Grover, Anuj (Advisor)
This report details the B.Tech Project (BTP) undertaken by Aryan Jain under the Entrepreneur ship Track at IIIT Delhi, focusing on the development of STEM Toy Kits and an IOT Activity Scoring Platform for the startup Robify. The project aims to revolutionise STEM education for school students aged 7 to 10 by creating engaging, hands-on learning experiences. The STEM Toy Kits, including a sound-responsive crawling spider and a human-sensing dog bot car, intro duce concepts like sensor technology and robotics. The IOT platform enhances these kits by automating activity scoring, providing real-time feedback, and fostering competition through leaderboards. Despite challenges such as hardware integration and performance optimisation, the project achieved functional prototypes and a robust digital platform. This report outlines the design, development, integration, challenges, and future directions of Robify’s innovative approach to STEM education.
2025-07-20T00:00:00ZAMR senseDindigallu, Hemanthhttp://repository.iiitd.edu.in/xmlui/handle/123456789/21642026-09-17T22:00:40Z2025-07-19T00:00:00ZAMR sense
Dindigallu, Hemanth
The AMR Sense Android application is a comprehensive platform designed to enhance awareness and management of antimicrobial resistance (AMR) through a user-centric mobile interface. De signed for various user groups, including individuals, ASHA workers, laboratories, and policymakers, the application facilitates the creation and authentication of secure accounts through mobile OTP and email-based verification. A Unity-based interactive game educates users about AMR through question-driven challenges, rewards, and badges to foster engagement. The app also incorporates forms to track antimicrobial drug usage and dosage, with plans to integrate Optical Character Recognition (OCR) for seamless prescription reading. Data is currently stored in Firebase Realtime Database, with a planned transition to a local PostgreSQL server to accommodate scalability. By combining gamification, data collection, and advanced technologies, AMR Sense aims to empower users with knowledge and tools to combat AMR effectively, contributing to public health initiatives. This report details the system architecture, implementation, challenges, and future enhancements of the AMR Sense platform.
2025-07-19T00:00:00ZSolving healthcare application of speech processingKumar, NiteenAgarwal, NiketBuduru, Arun Balaji (Advisor)http://repository.iiitd.edu.in/xmlui/handle/123456789/21162026-09-09T22:00:46Z2025-07-18T00:00:00ZSolving healthcare application of speech processing
Kumar, Niteen; Agarwal, Niket; Buduru, Arun Balaji (Advisor)
In order to improve human-computer interaction and create emotionally intelligent systems, au tomatic emotion classification from speech is essential. Large pre-trained audio language models (ALMs) have shown good generalization on a variety of tasks, but little is known about how well they perform on low-resource and multilingual datasets, particularly when code-switched or bilingual speech is involved. In this study, we examine the ability of cutting-edge pre-trained models to categorize emotions from a bilingual speech dataset that includes both Tamil and English utterances. To ensure balanced representation in both languages, our carefully selected dataset contains labeled audio samples from core emotion classes, including happy, sad, angry, and neutral. Using sophisticated ALMs such as Wav2Vec2, HuBERT, and Whisper, we extract fixed-length embeddings. We then assess these representations using Convolutional Neural Net works (CNNs) and Fully Connected Networks (FCNs). We investigate a dual-branch CNN archi tecture supplemented with a contrastive loss component to address intra-class language variance and inter-class emotion similarity. Our findings shed light on embedding-level language-agnostic emotion representation and demonstrate the potential of ALMs in robust emotion recognition, even in multilingual contexts.
2025-07-18T00:00:00ZComputational analysis of recipe structures and flavor networksBharti, PranjalBagler, Ganesh (Advisor)http://repository.iiitd.edu.in/xmlui/handle/123456789/21132026-09-08T22:00:47Z2025-01-01T00:00:00ZComputational analysis of recipe structures and flavor networks
Bharti, Pranjal; Bagler, Ganesh (Advisor)
This project explores how ingredients and flavor compounds interact across global cuisines using a data-driven approach. By leveraging structured datasets from RecipeDB and molecule-ingredient mappings, we analyze recipe structures, flavor-sharing indices, and statistical deviations from randomness. The study investigates regional trends in ingredient use and pairing logic, validated with customized Z-scores and delta metrics.
2025-01-01T00:00:00Z