dscripka/openWakeWord

Health v2 · maintenance only

Adoption not part of the score

2702 stars · 326 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

openWakeWord is an open-source Python library for detecting wake words (or phrases) in streaming audio, with pre-trained models for common phrases. It focuses on performance and simplicity, supporting both ONNX and TFLite inference runtimes.

Use cases

• detect a wake word to trigger a voice assistant
• add always-listening voice activation to a home automation setup
• run keyword spotting locally on a microphone stream
• train a custom wake word model for a specific phrase
• build voice-enabled applications without cloud speech APIs

When to choose

• you need lightweight, on-device wake word detection with minimal dependencies
• you want pre-trained models for common phrases that work in real-world environments
• you want to train a custom wake word model quickly using provided notebooks

When to avoid

• you need full speech-to-text transcription rather than wake word detection
• you require cloud-scale speaker verification or multi-language production support

Facets

library · maturity active

speech-recognition machine-learning audio-processing llm-inference speech-processing machine-learning python windows cross-platform wake-word-detection voice-assistant onnx tflite pre-trained-models keyword-spotting audio linux macos

Member repositories

For agents

markdown · JSON · MCP: product_card(name="dscripka/openWakeWord")