Oremi Device

Oremi Device captures microphone audio and connects it to Oremi services:

• Oremi Ohunerin detects wake words and environmental sounds.
• Oremi Andika transcribes speech after Ohunerin detects a wake word.
• Oremi Izwi or an OpenAI-compatible Edge-TTS server synthesizes spoken NLP responses when configured.
• MQTT exposes events, device controls, availability, and Home Assistant discovery entities.

The application monitors audio hardware while it runs. It can switch to another compatible microphone when the active device disappears and switch back when a preferred device becomes available again.

Documentation

All project documentation—including architecture details, configuration reference, and the Home Assistant integration guide—can be found at: https://demsking.gitlab.io/oremi-device/

Development

Oremi Device is built with Python and uses uv for dependency management. The project follows standard Python development practices with type hints, linting, and automated testing.

For detailed information on setting up your development environment, running tests, code style guidelines, and the pull request process, please refer to CONTRIBUTING.md.

Versioning

Oremi Device project adheres to Semantic Versioning (SemVer). Version numbers follow the MAJOR.MINOR.PATCH format:

• MAJOR version increments for incompatible API changes
• MINOR version increments for backward-compatible new functionality
• PATCH version increments for backward-compatible bug fixes

The current version can be found by running:

oremi-device --version

License

Copyright 2026 Sébastien Demanou. All Rights Reserved.

Licensed under the Apache License, Version 2.0. See the LICENSE for the complete license text.