Play sounds via MQTT (with Home Assistant autodiscovery). Multiple sound pools can be configured, each one will add a new switch to Home Assistant.
Sounds can either be played randomly (default) or with a custom ISoundSequenceProvider.
An example SoundSequenceProvider to tell the current time including the required sounds in german by Zabex is provided (see his microcontroller project to tell the time at https://www.zabex.de/frames/169zeitansage.html).
See "src/HomeAssistantUnifiLed/configs/appSettings.example.json" for a config example.
ffmpeg and ffplay has to be installed for audio playback. In the Dockerfile, this will be installed at build time automatically.
If you don't want to use this via Docker (or on Windows), make sure ffplay is available in PATH.
By default, the app looks for an configs/appSettings.json file or environment variables. An example config is provided in src/HomeAssistantSoundPlayer/configs/appSettings.example.json.
Make sure all submodules are checked out by either cloning this via git clone --recurse-submodules or running git submodule update --init.
Afterwards, a simple dotnet run --project src/HomeAssistantSoundPlayer/HomeAssistantSoundPlayer.csproj or Visual Studio build should be enough to get it up and running.
Use the docker-compose file or use some other way to run dotnet HomeAssistantSoundPlayer.dll and keep it running.