Skip to main content

A wake word detection system for Rhasspy

Project description

A wakeword detector for Rhasspy.

The Rhasspy project already has several wake word detectors to choose from. Unfortunately the existing wake word detectors fall in two categories: the not very good and the limited.

OpenWakeWordListener is sensitive, accurate, fast, and you can choose any wake word you like - you can even have multiple wake words or offload the wake word detection to another computer entirely. The trade-off is that OpenWakeWordListener requires more computing resources.

Configuring Rhasspy

  1. Open the Rhasspy web UI.
  2. Click the little cog icon in the left margin to open the settings.
  3. To the right of "MQTT", press the drop-down menu and select "External".
  4. To the right of "Wake Word" press the drop-down menu and select "Hermes MQTT".
  5. Click "Save Settings"

That's it! While you're in the settings, make note of which MQTT settings Rhasspy uses - you need to use the same settings for OpenWakeWordListener.

If you are running Rhasspy as a Home Assistant add-on, make sure you have also installed the Mosquitto MQTT broker add-on.

Running

To run the detector standalone: python src/openwakewordlistener/main.py

Give the -h command line switch for a list of command line arguments.

Dependencies

OpenWakeWordListener uses Silero VAD for speech detection and Whisper for speech decoding.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

openwakewordlistener-0.0.5.tar.gz (5.0 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

openwakewordlistener-0.0.5-py3-none-any.whl (6.7 kB view details)

Uploaded Python 3

File details

Details for the file openwakewordlistener-0.0.5.tar.gz.

File metadata

  • Download URL: openwakewordlistener-0.0.5.tar.gz
  • Upload date:
  • Size: 5.0 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.1.0 CPython/3.13.3

File hashes

Hashes for openwakewordlistener-0.0.5.tar.gz
Algorithm Hash digest
SHA256 994c0c2ae07f0aa93942a7d186805b2d92e28c37f12b151e333e9bd0ea4df1b4
MD5 fa5a5a3983224bab56dc42beffc0103b
BLAKE2b-256 a5a4c5e2ecafe9264bb4ead39844e9f8e2d7f8b998a7d37c3886d325d28ad368

See more details on using hashes here.

File details

Details for the file openwakewordlistener-0.0.5-py3-none-any.whl.

File metadata

File hashes

Hashes for openwakewordlistener-0.0.5-py3-none-any.whl
Algorithm Hash digest
SHA256 2f7eb17ad895c7acd95e59c2dc7c439192cac98bbb9df9214847ca6fb3de40be
MD5 c87b320ff16c01c5bcecb57f1984a56b
BLAKE2b-256 7f2479cac51b03c073e26d56ebff65448028ccd7467fec442b412b07afe9456a

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page