The virtual secretary, Lisa, is designed to interact naturally with users through text, voice, emotion, and a real-time animated avatar.

A fascinating demo website built with Node.js and Azure OpenAI o series/gpt4o-realtime/gpt4o series model. This project serves as a great starting point for developers who are interested in developing chatbot applications using JavaScript and Azure OpenAI API.

Enhancing Patient Care with OpenAI: A Blazor and Azure Speech AI Medical Assistant Web App

🎙️ Voice-to-text typing application powered by Azure Speech Services. Supports multiple languages with automatic language detection, mixed-language recognition, and real-time speech-to-text conversion. Perfect for multilingual users and accessibility needs.

Real-time AI voice conversation platform with WebSocket, supporting streaming STT/LLM/TTS

Documentação do uso das ferramentas Azure Speech Studio e Language Studio, focando na análise de fala e linguagem natural

AI-powered Microsoft Teams meeting assistant with voice interaction, conversation memory, and intelligent responses. Built with Azure OpenAI, Speech Services, and Bot Framework. Includes complete Bicep IaC for automated Azure deployment.

React Native / Expo app that scores Arabic pronunciation in real time using Azure Speech AI, with personalized coaching from OpenAI GPT-4o. Firebase Auth + Firestore sync, shipped to Google Play.

A low-token expressive affect layer for everyday AI tools, derived from Intrinsic Affect for AI.

Windows-first desktop dictation app with Azure Speech-to-Text, safe text injection, adaptive suggestions, and privacy controls.

CLI tool per generare verbali automatici da registrazioni audio di riunioni condominiali. Usa Azure Speech + Azure OpenAI.

Real-time speech translator using Azure Cognitive Services.

Turn a markdown file into a self-narrating web deck. Azure TTS + reveal.js on Cloudflare Workers.

Voice typing for any Windows app.

Production ready pipecat voice agent deployed to Azure Container App with FastAPI. Infratructure with Terraform.

Serverless backend for Lux. Single-router Vercel architecture hosting pronunciation scoring (Azure Speech), voice cloning (ElevenLabs), AI coaching (OpenAI), and Supabase-backed identity and data.

Browser-based voice AI platform for adult English learners. Pronunciation assessment, AI conversation practice, and learner-voice cloning. Vite + vanilla JS, ElevenLabs, Azure Speech, OpenAI Realtime API.

Windows text-to-speech app med Azure Neural TTS, systemtray-ikon, realtids-ordmarkering och screenshot OCR

Enable push-to-talk voice input on Windows with AI-powered text cleanup, punctuation, and seamless insertion into any application.

This project is a practical companion and code example for the article