S

swecha

Telugu Farmer Assistant is a free, AI-powered platform for farmers in Telangana and Andhra Pradesh. It provides crop disease diagnosis, soil-based crop planning, and real-time weather updates — all in Telugu language, with an offline-first design for accessibility.

AI/ML-based tool that predicts the likelihood of diabetes using medical parameters such as glucose, BMI, age, and blood pressure.

AI-powered platform for collecting and analyzing multilingual cultural narratives from global communities, built with Streamlit and Python.

The Image Captioning project is an AI-based system that automatically generates textual descriptions for images. It combines computer vision and natural language processing (NLP) techniques to understand the content of an image and produce a human-readable caption.

The system uses deep learning models, typically a Convolutional Neural Network (CNN) for extracting image features and a Recurrent Neural Network (RNN) or Transformer-based model for generating sentences. This enables applications such as assisting visually impaired users, enhancing image search engines, and automating social media content tagging.

BharatVivid deployment project for managing configurations, CI/CD pipelines, and automated deployments. Ensures smooth integration, testing, and release management within the Swecha ecosystem.

This is the Bilingual AI powered chatbot that I developed as a part of the Summer of AI internship task.

Hyderabad Food AI Assistant The Hyderabad Food AI Assistant is an intelligent and interactive tool designed to help users discover and get information about food in Hyderabad. It serves as a conversational assistant for exploring the city's diverse culinary landscape, providing details on restaurants, specific dishes, ratings, prices, and more.

The assistant is built with a hybrid search architecture to ensure both speed and flexibility. It first attempts to answer a user's query by searching a structured CSV file (hyderabad_menu_100.csv). This approach is highly efficient for specific, data-driven questions. If the structured search does not yield any results, the system intelligently falls back to a large language model (Flan-T5) to provide a more general, conversational response.

The application is built using the Streamlit framework, providing a clean and easy-to-use web interface for users to interact with the assistant. The core logic is encapsulated within the

FoodAssistant class, which manages the seamless transition between the CSV-based search and the LLM fallback. This architecture ensures that the assistant is always able to provide a helpful response, whether the information is contained in its local data or requires broader AI knowledge.

AI-powered chatbot application with React frontend and Node.js backend for interactive conversations

తెలంగాణం (Telanganam) 🪔

తెలంగాణం అనేది తెలంగాణ యొక్క అపారమైన సాంస్కృతిక, భాషా మరియు చారిత్రక వారసత్వాన్ని సేకరించడానికి, భద్రపరచడానికి మరియు లేబుల్ చేయడానికి రూపొందించబడిన ఒక బహిరంగ డిజిటల్ వేదిక. సమాజం అందించిన డేటా ద్వారా, ఈ ప్రాజెక్ట్ ప్రాంతానికి ప్రత్యేకంగా రూపొందించబడిన ఇండిక్ AI మోడళ్లను రూపొందించడం లక్ష్యంగా పెట్టుకుంది. జానపద కథలు, జానపద గీతాలు, పారంపారిక నిపుణులు, స్థానిక చరిత్ర మరియు భాషా మాండలికాలు వంటి వాటితో సహా తెలంగాణ ఆత్మను డిజిటల్‌గా భద్రపరచడానికి మాకు సహకరించండి.

Telanganam 🪔

Telanganam is an open digital platform designed to collect, preserve, and label the immense cultural, linguistic, and historical heritage of Telangana. Through community-contributed data, this project aims to empower the creation of Indic AI models specifically tailored for the region. Join us in digitally preserving the soul of Telangana, including folk tales, folk songs, traditional skills, local history, and linguistic dialects.

A cryptocurrency price prediction system using LSTM neural networks. Groups coins by market dynamics and forecasts future prices for investors and traders.

SwechaBot: A conversational AI chatbot built with Streamlit, Hugging Face Transformers, and Torch

A Telugu Proverbs Chatbot built with Streamlit and SQLite to explore and preserve traditional Telugu Proverbs.

Basha Kahani

Basha Kahani is a storytelling web application that allows users to create, share, and explore short stories in an engaging and user-friendly way.

🔹 Features:

Add and publish your own stories.

View stories shared by other users.

Interactive UI with smooth navigation.

Built with modern web technologies for performance and scalability.

🔹 Tech Stack:

Frontend: React.js (with TailwindCSS for styling)

Backend (if applicable): Node.js / Express.js (or API integration)

Database (if applicable): MongoDB / Firebase / JSON-based storage

🔹 Project Goals: The main goal of this project is to provide a creative platform where users can express themselves through stories. It is designed to be simple, lightweight, and easy to use, while also serving as a learning project to practice React, component-based design, and UI/UX development.

The English-to-Telugu Translator is a web-based application that allows users to easily translate text from English into Telugu. It leverages advanced language processing techniques to provide accurate and context-aware translations, helping users understand and communicate in Telugu more effectively.

Key Features:

Translate English text into Telugu instantly. User-friendly interface for easy input and output. Supports short phrases as well as long paragraphs. Can be extended to include text-to-speech for Telugu translations. Useful for students, professionals, and anyone learning Telugu.

Technology Stack:

Frontend: HTML, CSS, JavaScript (or Streamlit/Gradio for a simple UI) Backend: Python with translation libraries or APIs (e.g., Google Translate API, Indic NLP library) Optional: Speech synthesis for reading translated text aloud

Use Case: This tool can help language learners, travelers, or professionals who need quick and reliable English-to-Telugu translations, making communication more effective and accessible.

Doc Bot – AI Doctor Assistant is an interactive chatbot that provides doctor-style suggestions based on user-described symptoms. Built with Streamlit for a responsive and user-friendly interface, it integrates with the Dify.ai API to generate natural language responses. This project was developed as part of an internship at Swecha Telangana.

⚠️ Disclaimer: Doc Bot is for educational and informational purposes only and does not provide professional medical advice.

Local Proverbs Collector is an open-source, AI-powered application that helps preserve and celebrate traditional Indian proverbs across multiple languages and dialects. The app allows users to contribute proverbs through text or voice, automatically transcribing speech, translating into other languages, and optionally reading them aloud using Text-to-Speech.Optimized for rural and low-connectivity areas, it supports offline-friendly usage, light/dark mode, and a gamified experience with leaderboards, streaks, and voting.All data is stored in structured JSON format for easy research, analysis, and integration into AI language models ensuring India’s oral heritage is preserved for future generations.

Project Description An AI-powered Healthcare LLM designed to assist visually impaired (blind) individuals. The system provides accessible medical guidance through natural language processing, voice interaction, and context-aware support. This project aims to enhance healthcare accessibility, improve independence, and empower blind users by bridging the gap between medical knowledge and inclusivity.

A Streamlit-based music library application that allows users to browse, search, and stream audio tracks according to thier mood and also it is an otp sign in thing where it is integrated with swecha apis. Built with Python and Streamlit, designed for fast deployment.

AI Voice Generator: Converts text to natural-sounding speech using AI models.

name: civicvoice-ai-app description: AI-powered civic engagement tool that summarizes public input and highlights actionable insights for governments. tags:

civic-tech NLP public-engagement AI default_branch: main visibility: public license: AGPL-3.0-only