Home Assistant Voice support

The Home Assistant integration exposes Assist as a voice assistant entity in the Remote Two/3. See Using Voice Assistants for more information.

This feature was introduced in beta firmware version 2.8.1 and Home Assistant integration version 0.14.1.

See Home Assistant documentation on how to set up Assist in Home Assistant: https://www.home-assistant.io/voice_control/

Supported features

Important: Speech-to-text must be configured, otherwise the pipeline is not exposed as a profile.

• Each Assist pipeline is accessible with a voice assistant profile.
• The preferred pipeline is used by default if no profile is selected in the Remote Two/3.
• The intent response and action texts are shown in the Remote UI
• If Text-to-speech is configured, the returned audio response can be played back on the Remote.

A typical Assist pipeline configuration in Home Assistant with all features enabled:

Add Voice Assistant entity

To be able to use Assist on the Remote Two/3, the new voice assistant entity must be configured.

1. Open the Home Assistant integration in the web-configurator.

2. Click on the + button in the configured entities list to add new entities from Home Assistant:

   

3. Search for the Assist voice assistant entity, select it, and add it to the configured entities:

   

4. The entity will now appear in the configured entities list and will be available as a voice assistant in the settings menu and within activities.

   

Configure Assist as voice assistant

To use Assist on the Remote Two/3 as a voice assistant, it needs to be configured either as a global default voice assistant, or within an activity. The following steps show how to configure Assist as a global voice assistant.

1. Open Settings, Voice control in the web-configurator

   

2. Select the voice assistant by clicking on the selection box showing “None”, to bring up a list of voice assistants:

   

3. The Assist pipelines are shown as profiles. A language label shows the configured pipeline language and icons showing the supported features:
   

Icons:

	Transcription Always available. The spoken command is converted to text and displayed on the UI.
	Text Response Always available. The result of the command (e.g., “Lights turned on”) is displayed.
	Speech Response Only available if text-to-speech is configured in the pipeline. The assistant returns an audio reply, usually the spoken text response, which can be played on the Remote device.