Skip to content
Afa' Afa'

Project

Oremi Ohunerin Listener

Oremi Ohunerin Listener bridges Oremi Ohunerin's real-time audio detection with MQTT, enabling seamless integration into home automation systems like Home Assistant.

36 commits 1 contributor Updated Jun 2026

Overview

Oremi Ohunerin Listener is an implementation of Oremi Ohunerin, the real-time audio detection component of the Oremi Personal Assistant project. It connects to the Oremi Ohunerin WebSocket server, detects environmental sounds and wake words, and publishes the results to an MQTT broker.

This project is designed to integrate seamlessly with home automation systems like Home Assistant, IoT platforms, or any application that uses MQTT for communication. It is available as a Docker image for both linux/amd64 and linux/arm64 architectures.

Features

  • Real-Time Audio Detection: Detects environmental sounds and wake words in real time.
  • Customizable Detection: Configure specific wake words and sounds to detect using a features.json file.
  • MQTT Integration: Publishes detected sounds and wake words to an MQTT broker.
  • Home Assistant Compatibility: Easily integrates with Home Assistant for automations, notifications, and dashboards.
  • Customizable Threshold: Set a confidence score threshold for sound detection.
  • Cooldown Period: Avoids publishing the same sound repeatedly within a specified delay.
  • Docker Support: Available as a Docker image for easy deployment.
  • Cross-Platform: Supports both linux/amd64 and linux/arm64 architectures.

Technology

  • home assistant
  • mqtt
  • Python
  • sound detection
  • wakeword

Repository information

Primary language
Python
Created
January 19, 2025
Last activity
June 13, 2026

Activity

Daily commits on the default branch, across the whole project history.

Daily commit activity over the project's whole history. Each cell is one day; darker cells mean more commits.
Mon Wed Fri

Contributors

Search articles and projects

Type to filter articles and projects. Use the arrow keys to move through results and Enter to open one. Press Escape to close.