International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 12 Issue: 04 | Apr 2025
p-ISSN: 2395-0072
www.irjet.net
VIVI: A Personalized Virtual Assistant Sai Patil*1, Zoha Parkote*2, Chinmay Mirlekar*3, Susmita Mistry*4, Deepti Janjani*5, Sanjay Patil*6 *1,2,3,4 Student, Department of Artificial Intelligence and Data Science, Datta Meghe College of Engineering, Navi
Mumbai, Maharashtra, India
*5 Assistant Professor, Department of Artificial Intelligence and Data Science, Datta Meghe College of Engineering,
Navi Mumbai, Maharashtra, India
*6 Head of Department, Department of Artificial Intelligence and Data Science, Datta Meghe College of
Engineering, Navi Mumbai, Maharashtra, India ---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - The accelerated development of artificial intelligence (AI) has brought about virtual assistants that simplify human-computer interaction. However, current systems are still limited by inadequate personalization, the absence of advanced task automation, and inadequate contextual retention from one interaction to another, which hinders their ability to provide truly seamless and intelligent user experiences. This work introduces VIVI, a cutting-edge virtual assistant that seeks to overcome these boundaries by bringing forth a combination of leading-edge technologies such as Generative AI and Natural Language Processing (NLP) to render dynamic, context-dependent answers and execute sophisticated tasks effortlessly. The architecture of VIVI includes a smart workflow that starts with system boot-up and hotword detection, then proceeds to command identification through speech or text inputs. These commands fall into types like system controls, media actions, web interactions, or AIpowered tasks. By enabling advanced automation of tasks and multi-platform compatibility, VIVI provides higher user flexibility, providing not only functional commands but also proactive and personalized outputs. VIVI reimagines the power of virtual assistants through enhanced user interaction through intelligent decision-making complemented by a polished task accomplishment, positioning it as a very strong tool in personal and workspaces alike.
conversations. Unlike traditional virtual assistants, VIVI enables dynamic user-adaptive answers through Generative AI, and conversations become more interactive and personalized. VIVI also supports higher-order task automation, and it is possible to run complex multi-step workflows and system administration tasks. Its design ensures seamless interoperability with third-party applications, allowing users to efficiently manage both personal and work-related tasks. Moreover, VIVI’s userfriendly interface, supporting both voice and text inputs, ensures flexibility and ease of use across a wide range of devices and environments. This project also elaborates on the applications of VIVI in the future for smart homes, offices, and IoT devices, rendering it a tremendous advancement in virtual assistant technology. By addressing current challenges in personalization and automation, VIVI creates a new standard in context-aware, intelligent virtual assistants.
2. LITERATURE SURVEY Mekni, Mehdi. (2021) proposed an AI-based virtual assistant using conversational agents. The virtual assistant was designed to support multimodal communication, including text and voice-based interaction, ensuring a more intuitive user experience. The study explored the integration of Natural Language Processing (NLP) and Machine Learning (ML) techniques to enhance user interaction [1]. A. Sudhakar Reddy M, Vyshnavi, C.and Saumya designed a virtual assistant using NLP for speech and text interpretation and ML for better responses. It automated tasks like scheduling, emails, and smart home control, making user interactions easier and more efficient [2]. Baskaran, G., Raj, Harrish, Kumar, S., and Anand, R. (2021) developed an AI-powered virtual assistant for Windows OS. It used voice commands to manage applications via a command prompt and optimized system performance with user-defined models. Reinforcement learning improved adaptability based on user interactions [3].
Key Words: Generative AI, Virtual Assistant, Task Automation, Natural Language Processing, AI-powered Systems, Speech Recognition, Personalized User Experience.
1.INTRODUCTION VIVI addresses the shortcomings of existing virtual assistants, particularly in terms of personalization and task automation. VIVI, our AI-powered virtual assistant, aims to overcome these shortcomings by employing Generative AI, Natural Language Processing (NLP), and sophisticated automation techniques. With VIVI, users can expect personalized responses, contextual understanding, and smart task management across multiple platforms, giving a more natural and personalized experience.
Harshit Agrawal, Nivedita Singh, Gaurav Kumar, Dr. Diwakar Yagyasen, (2023) developed a Python-based voice assistant to execute Linux commands using voice input, reducing the need for keyboards and mice. The system improved
The key contributions of VIVI are that it supports keeping contextual knowledge consistent across multiple interactions, allowing for more relevant and coherent
© 2025, IRJET
|
Impact Factor value: 8.315
|
ISO 9001:2008 Certified Journal
|
Page 374