Can't get Atom Echo to work

Configuration Voice Assistant
1

Using RPi 4, Home Assistant Core 2023.11.2, Supervisor 2023.11.0, Operating System 11.1, Frontend 20231030.2

I used Microsoft Edge to install Voice Assistant on the Atom Echo. I used the instructions at

I set up the pipelines and the VA works when using the microphone on my laptop. When I try the Atom Echo, it does not work. I went to ESPHOME and clicked on “Edit” to see what was on the Echo. There was 17 lines of code, which looks like the barebones that you get when you set up a device in ESPHOME. Here is what was on the Echo…

substitutions:
  name: m5stack-atom-echo-88XXXX
  friendly_name: M5Stack Atom Echo 88XXXX
packages:
  m5stack.atom-echo-voice-assistant: github://esphome/firmware/voice-assistant/m5stack-atom-echo.yaml@main
esphome:
  name: ${name}
  name_add_mac_suffix: false
  friendly_name: ${friendly_name}
api:
  encryption:
    key: 65qj7KJ6OMXXXXXXXXXXXXXXXXXXXXs+/ZudeoCdC3tzk=


wifi:
  ssid: !secret wifi_ssid
  password: !secret wifi_password

Where was all the code like in the github yaml file? I then clicked on “Validate” and it showed me what looked like the 200 or so lines of code from github. I clicked on “Install” and a few minutes later it compiled and uploaded to the Echo. Yet when I clicked on “Edit” to see what was on the Echo, I only saw the original 17 lines of code!

I got the same results using Chrome. I am running Windows on my laptop. When I am chasing my tail like this it usually means that I am missing something embarrassingly basic. Can someone steer me right? My knowledge level is on the low side, yet I have been able to follow instructions for the last two years setting up Home Assistant.

2

Looks like you’ve got it set up on your Atom Echo. The 17 lines of code or so actually includes pulling in the code from github.

Is it added to the ESPHome integration? Go to Setting → Devices & services → click X Devices (where X is the number of devices currently being managed by the integration). Your Atom Echo should show up there and you should be able to see options like “Use Listen Light” and “Use wake word” as switches. If it’s not there, check your notifications and see if it can be added.

If it’s there already, then congrats, you’ve already set up your Atom Echo. You probably have a problem elsewhere. Since you can use Assist on your laptop, I’m going to assume your voice pipeline is correctly set up. My last guess is that you’re missing the wake word config in your Voice Assistant Settings. Go to Settings → Voice assistants → click on your Voice Assistant. Your conversation agent should be Home Assistant, Speech to text is faster-whisper, Text-to-speech is piper, Wake word engine is openwakeword. Pick a wake word like ok nabu since it’s been trained well.

If you’re missing any of the required options, go and add them as add-ons and verify they show up under the Wyoming Protocol integration.

If you’ve done all that and things still aren’t working, try reloading the add-ons (mine occasionally get hung up) and as a last resort, restart Home Assistant.

Hope this helps!

1 Like
3

You don’t say if this is your first ESPHome device. Have you set up your SSID and Password in secrets?

secrets

Click on the secrets icon and enter your WiFi credentials like this:

#WiFi credentials
wifi_ssid: mywifiname
wifi_password: mywifipassword
4

This is the third device that I am using in ESPHome and everything has been fine.

WHOA! I looked at my “Secrets” file, and the SSID specified there is different from the one that I configured the Echo to look for. I have several SSIDs floating around the house. I bet that is it.
Unfortunately, the HA unit is at a remote location, and I cannot get back there right away. From my original post, “When I am chasing my tail like this it usually means that I am missing something embarrassingly basic.” Well, yeah.
I will let you all know when I can get back to the cottage if that was it.

Many, many thanks.