Configure the microWakeWord model independently

[genehand]

Component name

micro_wake_word

Link to component documentation on our website

https://esphome.io/components/micro_wake_word/

Describe the enhancement

Currently we're able to specify multiple wake word models. When a voice_assistant is configured, a select component is also added to switch between them. I'd like to be able to configure the wake word model independently, when a voice_assistant is not part of the config.

Use cases

This is definitely an edge case. I made a custom flutter app for a tablet dashboard and voice assistant. Some wake-word options are out there, but nothing really viable for non-commercial use.

Using an Atom Echo running microWakeWord, it triggers the tablet app to start recording and run the rest of the assist pipeline. This is working great for me, but I found I need to add the voice_assistant: component to be able to configure the wake word model. That's fine as a workaround but ideally these components could be decoupled. Thx!

Anything else?

No response

[github-actions[bot]]

🏷️ I've automatically added the component: micro_wake_word label to help categorize this feature request.