An android client for OpenAI codex

"🌾 AgroGPT: An enterprise-grade AI advisor. Features voice/chat support in regional languages, built with FastAPI, React, and Docker. 🚀 Includes CI/CD automation and API security."

Website localization with Sarvam AI Translate

VivaAI is a next-generation AI-powered interview platform designed to streamline the hiring process through real-time communication and intelligent assessment. Built with cutting-edge technologies, VivaAI provides a seamless, professional interviewing experience for both recruiters and candidates.

🩺 Your in-house doctor - turning medical reports into easy-to-understand, mana local-language voice explanations.

Real-time speech translation for live presentation rooms. One QR, every language, interpreter-style latency. Speech-to-speech (STT to translate to TTS) over WebSocket, powered by Sarvam AI + optional OpenAI Realtime.

Helping Telugu NRI remain fully connected to family in India ❤️

Multilingual voice and text conversational AI agent that automates sales calls, contextually answers questions using RAG, and dynamically segregates leads into HOT/WARM/COLD pipelines in 10 Indian languages.

Modular real-time voice agent for Indian languages: cloud, local-English, and fully-offline Indic modes with VAD turn-taking and streaming TTS.

Redline is a voice-first crisis drills on a live incident, talk it down out loud, and get coached on every word once it's over

Local content-to-Obsidian pipeline. Paste any URL, get a structured markdown note.

AI-powered real-time voice agent for D2C e-commerce support — handles Hindi, English & Hinglish calls, detects frustration from voice tone, escalates intelligently to human agents. Built with FastAPI, OpenAI, Sarvam AI, LangChain + FAISS.

Speech-first Pi package powered by Sarvam AI

An AI powered resume analyzer and interview coach . It helps you quickly check your ATS score, get clear feedback on your resume, and practice interview questions made specifically for your profile.

Unofficial Go SDK for Sarvam AI - Chat, Speech-to-Text, Text-to-Speech, and more.

pipeline for building a curated TTS training dataset from youtube audio

India's Advanced Sovereign Multimodal AI Assistant Try the Web App here : https://aagni-ai.veercel.app

Hybrid voice dictation for Windows — types anywhere your cursor is. Choose between offline faster-whisper and Azure/Sarvam/Gemini cloud backends.

Vishwagyan: A Sovereign Multilingual AI Knowledge Engine for Bharat. Optimized for UPSC & high-stakes education using RAG & Indic LLMs.

A privacy-first, fully local Agentic AI legal assistant. Combines multimodal RAG (DeepSeek-R1 + Llama 3.2 Vision) via Telegram for zero-data-leakage contract analysis.