Voice cloning using Coqui Text-To-Speech (XTTS v2) with PyTorch and CUDA acceleration

Convert a markdown files into an audio narration using VCTK and Coqui-TT

A professional, high-fidelity AI Text-to-Speech (TTS) and Voice Cloning engine. Built using Coqui XTTS v2 for realistic, 100% offline vocal synthesis and custom voice cloning with privacy and performance at its core.

Stop paying monthly fees. High-fidelity voice cloning on your own GPU. No limits, no credits.

Local AI voice cloning with Coqui TTS XTTS-v2 - Docker-ready, GPU-accelerated

A fully offline, self-contained voice interaction system featuring speech-to-text, text-to-speech with voice cloning, and configurable wake word detection. Designed to run as a standalone service with comprehensive API endpoints.

Virtual AI Assistant – A scalable, real-time AI-powered avatar platform with Live2D, WebSockets, LLM integration, and TTS, designed for immersive experiences and future AR/VR extensions.

VoxBridge is an all-local speech pipeline that denoises audio, transcribes with Whisper, cleans transcripts via LLMs, and regenerates speech with Coqui, while exporting metrics to Prometheus/Grafana. Its modular backend and Streamlit UI let you swap models, process audio, and download outputs, all deployable via Docker Compose.

This is a proof of concept documentation for creating an audio book using free TTS tools

Setup and test coqui tts with local installation and docker image to setup a tts service

A mobile-first web app that reads academic papers aloud using a cloned voice, built for passive listening during moments where you want to engage your audio-sensory cognitive channels even while occupied with other task e.g long commutes and boring physical tasks.

Local offline multimodal story creation assistant. Unified SIS semantic schema + Docker microservices for reproducible generation & editing.

TTS (Text-to-Speech) Bahasa Indonesia menggunakan Coqui TTS dengan model VITS multi-speaker. Mendukung normalisasi teks khusus Indonesia dan 80+ speaker voice.

Voice Assistant using DistilGPT2 + Coqui TTS + Vosk (Offline STT)

Zero-shot voice cloning using XTTS v2. Upload a voice, type text, get audio

This repository is a comprehensive playground for audio experiments, exploring the intersection of modern AI and classic sound synthesis.

A multilingual AI voice agent that speaks, listens, and understands Bangla & English. Automates calls, confirms orders, and schedules appointments.

complete command-line pipeline for dataset preparation, fine-tuning, and inference with the XTTS v2 text-to-speech model.

A full-stack voice-to-voice AI agent built with React.js frontend and Flask backend. Users interact using voice, which is processed by an LLM (Google Gemini), and the response is spoken back.

Full Stack AI Agent working on locally deployed LLMs