Excited about proactive conversations with LLMs. But when creating automations with “Assist satellite ‘start conversation’” action, only “voice assistant preview edition” speakers show up. How can we add Atom Echo speakers to this capability?
It’s definitely not possible on current firmware.
1 Like
Should I petition ESPHome devs or do we have our own Atom Echo devs? I’d like this feature as well as ability to direct speaker output to any HA recognized media device (like google/amazon speakers). The workaround doesn’t work well.
Yep but when should this feature arrive to Atom Echo? in the release notes i see this:
As of this release, this feature is now available for ESPHome-based voice assistants
So its not true but why?
The announcement is about a specific device.
VPE uses modified components customized for it.
The common voice_assistant library has an older version without the necessary flags.
Wait for updates.
Your maybe right but then i can’t read 
what is your understanding of “available for ESPHome-based”
I know that i have to wait but i was confused.
Just out of curiosity: is this not just a question of copying the relevant code blocks from a VPE?
I would imagine it’s probably this bit of code here, but I have no idea how to integrate that into the code my own assistant (ESP32 S3) since the code seems to refer to home-assistant-voice-pe…
voice_kit:
i2c_id: internal_i2c
reset_pin: GPIO4
firmware:
url: https://github.com/esphome/voice-kit-xmos-firmware/releases/download/v1.3.1/ffva_v1.3.1_upgrade.bin
version: "1.3.1"
md5: 964635c5bf125529dab14a2472a15401
external_components:
- source:
type: git
url: https://github.com/esphome/home-assistant-voice-pe
ref: dev
components:
- micro_wake_word
- microphone
- nabu_microphone
- voice_assistant
- voice_kit
refresh: 0s