openwakeword
An open-source audio wake word (or phrase) detection framework with a focus on performance and simplicity
Decision gist · record as of 2026-08-14
Yes, if you need open-source wake-word detection with pre-trained models and low install friction. The permissive Apache 2.0 license and active repository are strengths. However, the aging maintenance status and lack of recent updates warrant caution if you require ongoing support or compatibility with the latest Python ecosystem. Suitable for prototyping and production use in voice-activated applications where you can tolerate a slower release cadence.AI-flagged interpretation of the facts on this page — verify before relying
Before you install
- Requires 16-bit 16kHz PCM audio input; on Windows, only onnxruntime inference is supported (tflite-runtime not available).
- Low install friction with a pure-Python wheel and six runtime dependencies.
- Maintenance is aging—last release was 2024-02-11, though the repository remains active and unarchived with 2662 stars.
License · maintenance · safety
permissive license (permissive) — Licensed under Apache 2.0 (permissive), allowing commercial and private use with minimal restrictions.
last release 2024-02-11 (915 days) · last repo commit 2025-12-30 · 2,662 stars
0 known vulnerabilities (OSV.dev, 2026-08-14) · 276,641 downloads/mo, #8,162 on PyPI
Alternatives
Verify before relying
pip install openwakeword
import openwakeword
from openwakeword.model import Model
openwakeword.utils.download_models()
model = Model()
prediction = model.predict(frame)- Whether the aging maintenance status will affect long-term compatibility with current Python and dependency versions.
- Real-world false-positive and false-reject rates for the included pre-trained models in production environments.
- Performance characteristics on specific hardware beyond the mentioned Raspberry Pi 3 example.
What it is and what it does
openWakeWord is an open-source framework for detecting wake words and phrases in audio streams. It provides pre-trained models for common trigger words that work in real-world noisy environments, and can be extended with custom models trained on synthetic speech. The library processes audio in 80 ms frames and returns confidence scores between 0 and 1, making it suitable for voice-activated applications like smart assistants, IoT devices, and web-based voice interfaces.
The package depends on onnxruntime and tflite-runtime for inference, plus scipy and scikit-learn for audio processing. It includes optional Speex noise suppression for Linux systems and a built-in Silero voice activity detection model to reduce false positives. The framework is designed to be lightweight enough for resource-constrained devices while remaining simple to integrate into Python applications.
Use it for
- Build a voice-activated smart home controller that listens for specific phrases to trigger actions.
- Add wake-word detection to a web application using the included streaming audio examples to process microphone input.
- Train a custom wake-word model for a specific application using the provided Google Colab notebook and synthetic speech generation.
- Reduce false positives in a voice interface by enabling voice activity detection (VAD) alongside wake-word scoring.
- Deploy multiple wake-word models simultaneously on edge devices for multi-trigger voice control.
- Batch-process large audio file collections to detect wake words using the bulk_predict utility with multiprocessing.
Worth the install?
AI-flagged interpretation of the facts on this page. Verify before relying on it.
Yes, if you need open-source wake-word detection with pre-trained models and low install friction.
The permissive Apache 2.0 license and active repository are strengths. However, the aging maintenance status and lack of recent updates warrant caution if you require ongoing support or compatibility with the latest Python ecosystem. Suitable for prototyping and production use in voice-activated applications where you can tolerate a slower release cadence.
Install
openwakeword on PyPI
Before you install
Low install friction with a pure-Python wheel and six runtime dependencies. Maintenance is aging—last release was 2024-02-11, though the repository remains active and unarchived with 2662 stars.
Requires 16-bit 16kHz PCM audio input; on Windows, only onnxruntime inference is supported (tflite-runtime not available).
License in practice
Licensed under Apache 2.0 (permissive), allowing commercial and private use with minimal restrictions.
Quickstart
pip install openwakeword
import openwakeword
from openwakeword.model import Model
openwakeword.utils.download_models()
model = Model()
prediction = model.predict(frame)Verify before relying
- Whether the aging maintenance status will affect long-term compatibility with current Python and dependency versions.
- Real-world false-positive and false-reject rates for the included pre-trained models in production environments.
- Performance characteristics on specific hardware beyond the mentioned Raspberry Pi 3 example.
Package facts
| License | permissive license permissive |
| Python support | Supports the current Python release >=3.7 |
| Install friction | Low. Pure-Python wheel |
| Runtime dependencies | 6 packagesonnxruntimetqdmscipyscikit-learnrequeststflite-runtime |
| Maintenance | Aging 915 days since the last release |
| Last repo commit | |
| First released | |
| Downloads | 276,641 / month, #8,162 on PyPI 30-day window, as of 2026-08-14 |
| Known vulnerabilities | None known OSV.dev, checked 2026-08-14 |
| Classifiers | License :: OSI Approved :: Apache Software LicenseOperating System :: OS IndependentProgramming Language :: Python :: 3 |
Evidence: openwakeword-0.6.0-py3-none-any.whl
Tags
Let your AI agent find packages like this
Example. Real query, live index.
You found this page by searching. An agent finds it by wishing: SkillFed indexes 14,416 PyPI packages by what they can do, searchable in plain language.
wish › “voice activation audio”
- openwakewordopenWakeWord detects wake words and phrases in audio streams using…
- pvporcupinePorcupine is a lightweight wake word detection engine that identifies…
- ResemblyzerResemblyzer generates a 256-value embedding that summarizes voice…
Give your agent the search over MCP, or paste the wish link into any chat.
More Artificial Intelligence packages
LiteLLM provides a unified Python interface to call 100+ LLM providers (OpenAI, Anthropic, Gemini, Bedrock, Azure, and others) using OpenAI-compatible API format, available as both a Python SDK and a self-hosted AI Gateway proxy server.
Install it if you need to work with multiple LLM providers or want to centralize LLM routing in your organization.
Client library and CLI tool for downloading, uploading, and managing models, datasets, and repositories on the Hugging Face Hub platform.
Install it if you work with Hugging Face Hub models or datasets.
LangChain provides a framework for building agents and LLM-powered applications by composing language models, tools, and memory through a unified API that abstracts over multiple model providers.
hf-xet provides chunk-based deduplication and efficient file transfer for the Hugging Face Hub, enabling faster uploads and downloads of large files with local disk caching.
Tokenizers converts raw text into token sequences for NLP models, with support for training custom vocabularies and using pre-built tokenizers (BPE, WordPiece) optimized for speed via Rust.
Transformers provides a unified framework for loading, fine-tuning, and running state-of-the-art pretrained models across text, vision, audio, video, and multimodal tasks using PyTorch, JAX, or TensorFlow.
Install it if you need to run or train any transformer-based model for NLP, vision, audio, or multimodal tasks.
See also pvporcupine · realtimestt · wyoming · silero-vad · pymicro-vad · webrtcvad-wheels · webrtcvad · pocketsphinx · whisper-timestamped · pyrnnoise