← All stories
● Covered by 1 source · 1 reportLow impact1 neutral

Platypush Voice Assistant Pipeline Updated with microWakeWord for Improved Detection

🔄 Updated 2d ago
New to BrevFeed? We gather this story from every outlet covering it into one summary — ranked by real-world impact, not just the latest headline — so you never miss what matters. What is BrevFeed? →

Key points

  • Replaced `assistant.openwakeword` with `assistant.microwakeword` plugin.
  • New plugin uses microWakeWord, also used by ESPHome and Home Assistant Voice.
  • Aims to reduce false positives/negatives in wake word detection.
  • microWakeWord models are actively maintained and ship with the package.

Pipeline Reconstruction

A developer has updated their self-hosted voice assistant pipeline built within Platypush. This rebuild focuses on improving the accuracy and robustness of wake word detection, a critical component for voice-activated systems.

Switching Wake Word Engines

The previous implementation used `assistant.openwakeword`, which wrapped the openWakeWord project. This setup resulted in high rates of false positives (triggering on background noise) and false negatives (failing to detect the hotword). The developer noted that openWakeWord's pre-trained models were from 2023 and primarily trained on synthetic speech.

The new pipeline integrates a custom `assistant.microwakeword` plugin, built on top of microWakeWord via `pymicro-wakeword` Python bindings. microWakeWord is also utilized by ESPHome and Home Assistant Voice, and its models are actively maintained. The package includes built-in models such as "okay_nabu", "hey_jarvis", "hey_mycroft", and "alexa", eliminating the need for separate downloads.

Improved Detection and Model Maintenance

The transition to microWakeWord is intended to provide more reliable wake word detection, addressing the previous issues with false triggers and missed detections, particularly in noisy environments or with less sensitive microphones. The active maintenance of microWakeWord's models, along with support for custom models, contributes to the system's improved performance. The new plugin maintains the same API as its predecessor, ensuring compatibility with existing configurations.

✨ This summary was generated by AI from the outlets' reporting listed below. It is not independently verified and may contain errors — check the original sources. How BrevFeed works →

The daily brief

One email each morning: the day's tech stories, clustered across outlets and summarized. No account needed.

One email a day. Unsubscribe in one click, any time.

Today's brief

Spend a few minutes, get the whole day. Every topic's top stories in one hands-free rundown — listen, watch, or read the transcript.

~4 min · 3 stories · Nov 21

▶ Play today's brief Listen on Spotify

New every morning, and the back catalogue is archived by date.

Reporting from

A developer rebuilt their self-hosted voice assistant pipeline in Platypush, replacing openWakeWord with a new `assistant.microwakeword` plugin. This change addresses issues with false positives and negatives experienced with the previous setup, aiming for more robust wake word detection.