Companion repository for the Code with Antonio video course · Watch on YouTube.
Build a JARVIS Voice AI Agent for Free with Python (100% Local): it hears with whisper.cpp, thinks with Ollama and talks with Kokoro.
A voice assistant that listens with whisper.cpp, thinks with a local model in Ollama and talks back with Kokoro in a British voice, with a personality, memory and Python skills for what the model can't know.
| # | Lesson | Video | Folder |
|---|---|---|---|
| 1 | What is a voice AI agent? | 0:00 | lessons/01-what-is-a-voice-ai-agent |
| 2 | The ears: whisper.cpp | 2:45 | lessons/02-the-ears-whisper-cpp |
| 3 | The mouth: Kokoro | 5:05 | lessons/03-the-mouth-kokoro |
| 4 | Ears in Python | 8:24 | lessons/04-ears-in-python |
| 5 | The brain: a local model | 10:25 | lessons/05-the-brain-a-local-model |
| 6 | Put it together: the loop | 13:27 | lessons/06-put-it-together-the-loop |
| 7 | Skills: what the model can't know | 15:47 | lessons/07-skills-what-the-model-can-t-know |
- whisper.cpp v1.9.4 · Ollama 0.35.0 + qwen3.5:4b · kokoro-onnx 0.6.1 (Kokoro-82M v1.0) · ollama-python 0.6.3 · sounddevice 0.5.6 · uv 0.12 · Python 3.12 · Ubuntu (Linux), CPU only
- Python 3.12+ and uv
cd app
uv sync
# whisper.cpp next to this folder (lesson 2): cd .. && git clone https://github.com/ggml-org/whisper.cpp.git && cd whisper.cpp && sh ./models/download-ggml-model.sh base.en && cmake -B build && cmake --build build -j --config Release
wget https://github.com/thewh1teagle/kokoro-onnx/releases/download/model-files-v1.1/kokoro-v1.0.onnx
wget https://github.com/thewh1teagle/kokoro-onnx/releases/download/model-files-v1.1/voices-v1.0.bin
ollama pull qwen3.5:4b
uv run demo_voice.py # optional: demo questions as WAV files
uv run main.py # press Enter and speak (or: uv run main.py demo/01.wav demo/02.wav)README.md this file
lessons/ one folder per lesson: goal, timestamp, commands and files exactly as in the video
app/ the final working project
assets/ thumbnail
MIT — use it for anything, credit appreciated.
Generated on 2026-10-02 from the course scripts. Code with Antonio · AI tutor.