3.5 KiB
🎙️ microWakeWord Nvidia Trainer & Recorder
Train microWakeWord detection models using a simple web-based recorder + trainer UI, packaged in a Docker container.
No Jupyter notebooks required. No manual cell execution. Just record your voice (optional) and train.
🚀 Quick Start
1️⃣ Pull the Docker Image
docker pull ghcr.io/tatertotterson/microwakeword:latest
2️⃣ Run the Container
docker run --rm -it \
--gpus all \
-p 8888:8888 \
-v $(pwd):/data \
ghcr.io/tatertotterson/microwakeword:latest
What these flags do:
--gpus all→ Enables GPU acceleration-p 8888:8888→ Exposes the Recorder + Trainer WebUI-v $(pwd):/data→ Persists all models, datasets, and cache
3️⃣ Open the Recorder WebUI
Open your browser and go to:
You’ll see the microWakeWord Recorder & Trainer UI.
🎤 Recording Voice Samples (Optional)
Personal voice recordings are optional.
- You may record your own voice for better accuracy
- Or simply click “Train” without recording anything
If no recordings are present, training will proceed using synthetic TTS samples only.
Remote systems (important)
If you are running this on a remote PC / server, browser-based recording will not work unless:
- You use a reverse proxy (HTTPS + mic permissions), or
- You access the UI via localhost on the same machine
Training itself works fine remotely — only recording requires local microphone access.
🧠 Training Behavior (Important Notes)
⏬ First training run
The first time you click Train, the system will download large training datasets (background noise, speech corpora, etc.).
- This can take several minutes
- This happens only once
- Data is cached inside
/data
You will NOT need to download these again unless you delete /data.
🔁 Re-training is safe and incremental
- You can train multiple wake words back-to-back
- You do NOT need to clear any folders between runs
- Old models are preserved in timestamped output directories
- All required cleanup and reuse logic is handled automatically
📦 Output Files
When training completes, you’ll get:
<wake_word>.tflite– quantized streaming model<wake_word>.json– ESPHome-compatible metadata
Both are saved under:
/data/output/
Each run is placed in its own timestamped folder.
🎤 Optional: Personal Voice Samples (Advanced)
If you record personal samples:
- They are automatically augmented
- They are up-weighted during training
- This significantly improves real-world accuracy
No configuration required — detection is automatic.
🔄 Resetting Everything (Optional)
If you want a completely clean slate:
Delete the /data folder
Then restart the container.
⚠️ This will:
- Remove cached datasets
- Require re-downloading training data
- Delete trained models
🙌 Credits
Built on top of the excellent
https://github.com/kahrendt/microWakeWord
Huge thanks to the original authors ❤️