ollama for deaf family member.

[peak plume]

I have ollama with agent llama3.2 and I would like to create an automation that reacts to each response and allows me to put the response in an input_text. I searched, but I haven't found a solution other than using {{ response.speech.plain.speech }} but it doesn't work. All this because I have a deaf family member and I would like him to read the responses. Thanks

[peak lotus]

What are you using as the voice satellite?

[peak plume]

i use two satellite, assist_satellite.vaca View Assist Companion App and M5Stack Atom Echo

[peak lotus]

I don't know about VACA, but on the atom echo you could potentially capture the response and send it somewhere as ESPHome gets the response test.

[peak plume]

Thanks for the reply. I'd also like to know if there's any way to read this attribute (locally processed). I'd need this for an automation that shouldn't happen if the attribute is false.
like this

{{ state_attr('conversation.llama3_2','processed_locally') }}

[stone wraith]

View Assist Companion App
try this project, it allows you to enable display of response text https://github.com/knoop7/Ava