A deep learning text classification system fine-tuned on BERT (bert-base-uncased) to identify signs of depression in written text. Features an automated training pipeline using Hugging Face's Transformers and an interactive CLI for real-time inference.

🧠 Recognize emotions from text, audio, and video to generate adaptive CBT-style responses for digital mental health support.

The Hotline Suicide Help Assistant, code-named Mehak, is an AI-powered crisis intervention prototype designed to provide immediate, warm, and professional support to individuals in distress. Leveraging the OpenAI Realtime API, Mehak offers a low-latency voice-to-voice experience, ensuring that those in need aren't left waiting.

AI-powered emotional state detection and guidance system using NLP, behavioral inputs, and machine learning with a live Streamlit demo.

Research prototype for structured psychiatric interviewing with LLM workflow orchestration, schema-governed outputs, and replayable evidence logging.

Mulit-Agent Setup that helps you recover from any emotional damage and helps you go back to routine and best form of yourself.

Transformer-based AI Safety Intelligence System for multi-label psychological distress, toxicity, and manipulation detection.

AI-powered support companions for anxiety, pain, dementia, menopause, autism, and life transitions. Empathetic guidance, therapeutic prompts, and 24/7 relief — built for real people.

Applied multimodal ML system combining text, audio, and visual emotion recognition with CBT-aligned response generation using late fusion.

State-based AI Agent for suicide risk detection and response using LangGraph and Python rule engine.

🚀 Space exploration platform for students & researchers + real-time AI emotion wellbeing companion ( SIH Problem Statement). Detects facial emotions via webcam, responds with empathy in 9 Indian languages using Gemini AI. Built with Python, Flask, DeepFace, PHP, MySQL & WebSocket. Open source. Made in BHARAT