Mind Guard - AI Mental Health Assistant
Nikhil Kumar Singh, Manoj M Naik, Prajna P Hegde, Soujanya M, Mr. Nithin H V
Paper Contents
Abstract
Mental health is now one of the biggest issues people face in the twenty-first century. It affects people of all ages, jobs, and places. According to the World Health Organization, almost one out of every eight people worldwide live with a mental health problem, like stress, anxiety, depression, or trouble sleeping. Even though more people are aware of these issues, its still hard to get the right kind of mental health help on time. This is because of high costs, not enough trained professionals, and the stigma that still exists around mental health. Even traditional therapy, which can be really helpful, often doesnt reach people when they need it most, leaving many without proper support. This paper talks about Mind Guard, an AI powered mental health assistant that aims to fill this gap by giving empathetic, voice-based, and privacy-focused help. Unlike regular text-based chatbots, Mind Guard uses natural speech-to-speech interaction, so users can have conversations that feel more human and emotionally comforting. Mind Guard isnt meant to replace real therapy, but to help with everyday mental health by offering tools for prevention and self-care. It gives features like breathing exercises, grounding techniques, and guided journaling. It also suggests relaxing activities, such as listening to music, based on the users preferences. People can choose their preferred language (English, Hindi, Kannada, Tamil), how the assistant talks, and what issues they want to focus on like stress, anxiety, or sleep. This helps make the platform more inclusive and easier to use for different groups of people. The system also cares about keeping user data private and secure. It stores conversation history safely, so users can look back on past talks while keeping everything confidential.
Copyright
Copyright © 2025 Nikhil Kumar Singh, Manoj M Naik, Prajna P Hegde, Soujanya M, Mr. Nithin H V. This is an open access article distributed under the Creative Commons Attribution License.