Wyoming Piper docker container with GPU support for Home-Assistant

This project combines embedded system and AI inference to create an end-to-end conversational experience.

Conversational AI, local, low-latency voice assistant for Raspberry Pi 5 with LED, gesture control, and streaming LLM replies.

A voice assistant with local LLM as a backend

(wip) python command-line Text-to-Speech (TTS) tool esp. for German, leveraging numerous endpoints like orpheus, piper, outetts, kokoro, csm, edge, coqui, kartoffelbox, etc

Quick integration of Piper TTS (super lightweight, high-quality model) with LiveKit

A simple, containerized web API for real-time text-to-speech (TTS) synthesis using Piper, powered by FastAPI.

⚙️ Isabella – a full-stack 🚀 conversational system built on FastAPI ✨ featuring 🎭 ML-powered emotion detection microservices (RoBERTa/Deep Learning) embedded into LLM, 🎤 advanced TTS generation pipeline, 🧠 memory-aware context handling,📊 rich logs + analytics , and 🔐 a dynamic persona delivering consistent, signature responses.

PyPiperTTS is a Python library that provides a simple and intuitive interface to the Piper text-to-speech system.

A simple, hands-free Python voice assistant that runs 100% locally. This script uses openwakeword for wakeword detection, webrtcvad for silence detection, OpenAI's Whisper for transcription, and Ollama for generative AI responses.

Simple Python Docker exposing an API using Nemo to perform text normalization using NVidia NeMo framework.

Python Text-to-Speech toolkit (multi-engine) with FastAPI, CLI and Telegram integration. Featuring first-class Persian support.

A command-line Python script to generate high-quality speech from text using the Piper TTS engine

A bare bones voice mcp server using piper

A fully functional 100% offline voice assistant with multi-language support.

A powerful Anki audio add-on with a 3-tier playback system: prioritizes your local audio dictionary, falls back to online gTTS, and finally to offline Piper TTS

An offline Text-To-Speech service you can host at home

Clipboard monitor that converts copied text to speech (TTS) using configurable engines (Edge-TTS, Piper, SAPI). Streams audio locally.

A project for the toxic commentator sitting next to you, commenting on your game.

A multilingual, AI-powered voice agent designed to make welfare payments and government schemes instantly verifiable and accessible for rural and semi-urban citizens. It reduces the economic and social burden of repeated bank trips by providing a simple, local, and verifiable way to check financial status, withdrawals, and scheme eligibility.