AI Voice companion system

AI Voice Companion System

— Local AI That Talks Back The AI Voice Companion System connects Ollama, Whisper.cpp, and Piper TTS into one simple local interface — letting you talk to your AI and hear it respond in real time, fully offline. 

How It Works

Whisper.cpp handles speech-to-text (your voice → text). Ollama processes that text through your chosen local AI model. Piper TTS turns the AI’s response back into speech. Everything runs locally — no cloud, no data sharing, no subscriptions. 

Requirements

Ollama AI installed on your PC At least one Ollama model downloaded (e.g. Gemma, Llama, Mistral, etc.) The included ZIP file contains preconfigured builds of Whisper.cpp and Piper TTS, which are free and open-source tools. 

Credits: Whisper.cpp by Georgi Gerganov & Piper TTS by Rhasspy

⚙️ Key Features

✅ 100% local — your data never leaves your machine
✅ Simple GUI for easy setup and control
✅ Works with any Ollama-based model
✅ Lightweight, fast, and low resource usage (Based on models used)

💡 Use Cases

• Create your own offline AI assistant

• Build a streaming or VTuber companion

• Experiment with speech-to-text and text-to-speech pipelines

• Integrate local AI into personal projects