“小鸭同学” wake word model for ESP32 and ESPHome
microWakeWord model for the wake word “小鸭同学”, trained on microwakeword.com from synthetic speech (Chinese). It runs on-device on ESP32-S3 boards with ESPHome's micro_wake_word component, including the Home Assistant Voice Preview Edition, with no cloud involved.
Model facts
- Wake word
- “小鸭同学”
- Language of the training voices
- Chinese
- Trained
- 2026-08-25 · microWakeWord v2
- Training steps
- 20,000
- Recall, clean speech
- 100.0%
- Recall, with background noise
- 99.8%
- False activations on the benchmark set
- 1.51 per hour at probability cutoff 0.5
- Sliding window
- 5
- File
- quantized INT8 TFLite, 59 KB, with a micro_wake_word manifest
- Downloads
- 23
Add it to your ESPHome device
ESPHome fetches the model straight from this site when it compiles the firmware (ESPHome 2024.7 or newer); nothing to download.
micro_wake_word:
models:
- model: https://microwakeword.com/api/models/2612/manifest.jsonSign in on this page to download the ZIP (TFLite + manifest) for offline use. Models are licensed for personal and non-commercial use by default (commercial license).
Similar wake words in the library
- “喂 V 仔” — 99.0% recall
- “哈密瓜.” — 95.6% recall
- “小尚小尚” — 100.0% recall
- “小聆小聆” — 98.6% recall
- “哈喽小赫” — 99.6% recall
- “hi lina” — 85.6% recall
Custom wake words are trained in the Training Center; the library lists every public model with the same measurements.