https://repository.iiitd.edu.in:443/xmlui The DSpace digital repository system captures, stores, indexes, preserves, and distributes digital research material. Tue, 08 Sep 2026 04:25:24 GMT 2026-09-08T04:25:24Z http://repository.iiitd.edu.in/xmlui/handle/123456789/2110 Distributed communication system Gaba, Divyansh; Dabas, Deepanshu; Singh, Pushpendra (Advisor) This project is dedicated to developing a robust chat application akin to WhatsApp, with a focus on leveraging RabbitMQ for efficient message queuing and Android development for creating an intuitive user interface. The primary objective is to achieve real-time messaging capabilities while integrating advanced features such as a comprehensive monitoring dashboard. This dashboard aims to provide users with insights into message statuses, including views and responses, catering to research and business needs. The project addresses several key challenges anticipated in modern communication applications. Ensuring message reliability and scalability in a real-time messaging environment is crucial, and the project aims to achieve this through a robust backend infrastructure utilizing RabbitMQ for message queuing. The goal is to ensure seamless message delivery even in high-demand scenarios. Security is another critical aspect that the project aims to address comprehensively. The implementation of robust encryption protocols and authentication mechanisms will be crucial in safeguarding user data and communication channels, ensuring end-to-end security and user privacy. One of the innovative aspects of this project is the integration of a micro-chat service, which enhances communication capabilities within other applications. For instance, in the healthcare sector, integrating our chat API into a medical mobile app enables secure doctor-patient communication. Our monitoring dashboard, coupled with efficient message queuing algorithms, can track patient interactions, ensuring timely responses and improved healthcare delivery. Through ongoing research and development efforts, the project aims to discover and implement efficient message queuing algorithms to optimize message delivery, reduce latency, and enhance user experience. The monitoring dashboard, powered by information retrieval techniques, is expected to provide actionable insights into communication patterns, aiding businesses in optimizing their communication strategies. The conclusion drawn from this ongoing project is the potential for successfully integrating complex technologies to deliver a seamless and secure messaging experience. The anticipated RabbitMQ-based backend will be instrumental in ensuring message reliability and scalability, essential aspects for modern messaging applications. The incorporation of advanced security measures is also expected to address growing concerns regarding data privacy and security in messaging platforms. The monitoring dashboard, although still in development, is anticipated to emerge as a stand- out feature, offering valuable analytics and insights that can be leveraged by businesses and researchers alike. This ongoing project lays the foundation for further advancements in communication technologies, emphasizing the importance of reliability, security, analytics, and the integration of micro-chat services in modern chat applications. Wed, 27 Nov 2024 00:00:00 GMT http://repository.iiitd.edu.in/xmlui/handle/123456789/2110 2024-11-27T00:00:00Z http://repository.iiitd.edu.in/xmlui/handle/123456789/2109 Speech processing and its applications for low resource languages Ekbote, Vijval; Gupta, Anubha (Advisor) Voice cloning (VC) has emerged as a transformative technology, enabling personalized Human Computer Interaction and enriching user experiences. However, the performance, efficiency, and environmental impact of these models require rigorous assessment and benchmarking. Current benchmarking techniques generally rely on the Mean Opinion Score (MOS), which is highly sub jective and impacted by individual biases. Therefore, this work proposes an automated framework, namely GreenVoice, for quantitative benchmarking of VC models using speaker verification. Green Voice is used to comprehensively benchmark four open-source state-of-the-art (SOTA) VC models based on the naturalness of their generated voice clones, costs, and the environmental impact while inferencing. GreenVoice has been used to investigate the performance of these models on out of-domain test sets to advance the development of inclusive models. Inclusive VC models show equitable performance for users from different demographics, including genders, accents, and low resource languages. Empirical results demonstrate a significant performance degradation of these models on out-of-domain datasets. Our findings reveal a higher naturalness in the generated clones for male speakers than female speakers. Additionally, we emphasize the importance of reporting the environmental impacts, such as carbon emissions, of using large AI models. With the view of having sustainable and inclusive VC technologies, we encourage researchers to follow the Green AI practices and work towards developing environment-friendly and robust speech processing models. Building on this, we further propose a GAN-like framework for fine-tuning any voice cloning model, and present some preliminary results for one of the models. Wed, 27 Nov 2024 00:00:00 GMT http://repository.iiitd.edu.in/xmlui/handle/123456789/2109 2024-11-27T00:00:00Z http://repository.iiitd.edu.in/xmlui/handle/123456789/2108 Digital humans in physical activity Singh, Asa; Sharma, Ojaswa (Advisor) In this work, we build on last semester’s work to improve how we give personalized exercise feedback using 3D human pose data. We start with two methods we already implemented: motion retargeting, which maps a therapist’s movements onto a patient’s body shape using SMPL, and shape-invariant embeddings trained with contrastive learning to focus on motion rather than body differences. This semester, we added two new components. First, a RotJoints alignment pipeline that converts SMPL poses into Euler angles and normalizes them, then ap plies full-sequence Dynamic Time Warping and an improved subsequence DTW for more flexible temporal matching. Second, a lightweight transformer model trained on the MinT dataset to predict muscle activations from SMPL pose sequences. Qualitative examples show that our subsequence alignment yields smoother synchronization compared to the basic DTW approach, and our muscle-activation model produces plausible activation patterns. Finally, we propose combining these kinematic and physiological cues into a single feedback system for rehabilita tion and fitness applications. Wed, 23 Jul 2025 00:00:00 GMT http://repository.iiitd.edu.in/xmlui/handle/123456789/2108 2025-07-23T00:00:00Z http://repository.iiitd.edu.in/xmlui/handle/123456789/2107 Splanet (Telecalli)- revolutionizing AI-powered cold calling and customer support Kapoor, Janesh; Grover, Anuj (Advisor) The competitive nature of business outreach necessitates efficient and personalized customer interactions. "Telecalli" is an innovative AI-driven platform designed to address these needs by automating cold-calling workflows, providing human-like conversational capabilities, and integrating seamless follow-up scheduling. This platform empowers businesses to scale their outreach while maintaining the quality of interactions. Using cutting-edge technologies like Google Gemini, Twilio, and React.js, Telecalli bridges the gap between automation and personalization. The report delves into the development process, tools utilized, challenges faced, and the impactful results achieved through this transformative project. Wed, 27 Nov 2024 00:00:00 GMT http://repository.iiitd.edu.in/xmlui/handle/123456789/2107 2024-11-27T00:00:00Z