#STT and TTS sensor

1 messages · Page 1 of 1 (latest)

fallen delta
#

Hello, Does anyone know is there a way to create two template sensors that would capture both the SST and TTS that the Voice PE heard and responded with. This would be similar, to the sensors in View Assist Companion App. The data is in the Debug Assistant so I figured that has to be a way to capture it. I just don't know how.

livid remnant
fallen nacelle
#

It is unclear what you are asking. That might be a skill issue on my part, but I’m interpreting this as you want different wake words. Is that what you are asking?

If so you can assign different wake words from the voice PE device view.

The other interpretation is that you want something that would trigger either before or after the voice PE responds to you. If that is the case you could set an automation to trigger when the state for the PE switches from responding to idle and have the action part of the automation use your template to do something.

fallen delta
#

Hello all let me try to clarify the use-case. It would be helpful when one makes a request to have a sensor to see what the device actually heard. As we all know sometime our pronunciation is not always the same and what we say is not always what is heard by the voice assistant.

Also there are times when you may want to monitor the kind of request or commands being given, especially when you have kids.

Alexa Media Player (HACS by Keith Taylor) has a sensor called last_called_summary that captured the last command give to the Alexa device.

Most recently View Assist Companion App has two sensors that captures STT and TTS. See the provided screenshots.

All I'm asking is can we have something similar, either exposed to us by the Home Assistant development team, or is it possible to create a template sensor that captures that info. Currently, if you go to Settings-->Voice Assistants-->Click the 3-dots-->Select Debug. You will see the STT and TTS info.

Thanks

image.png image.png
fallen delta
#

STT and TTS sensor