Original yaml of the 13$ atom echo voice assist?

Hi all,

can the code of the original atom echo voice assist be found anywhere?
I would want to add the device to the esphome dashboard after installed using the “website”:

BR

There is a link in the page to media-players/m5stack-atom-echo.yaml at 01db8a9340d0d405da79e756266339e77930c243 · esphome/media-players · GitHub

1 Like

Thanks for the info! Found that as well… strange is: using the automatic install everything works just fine. Using this yaml code the led does not light up when the button is pressed and there is no feedback given through the speaker…

1 Like

Actually I think it is this one

1 Like

Thanks … will try that as well :slight_smile:

Update: Thats its!

I have these little devices and they work for talking and sending commands into Home Assistant, however, I feel like the device should be saying something back after each command is given. Like, I say: “Turn off the office light” and it should respond with “turned off light” but it never says anything back. I think it is supposed to based on the Voice Assistant debug screen but even clicking on the “Play audio” button yields an error.

It should say something back, right?

EDIT: By the way, I know that these little guys work as I can use the service to have Home Assistant and voice to it and it speaks correctly. So, I think this is a bug in the Voice Assistant software at the moment.

did you ever get this to work? I am not getting any audio feedback either. Had to make changes to the code to make it build and deploy first. Confirmed speaker works using the TTS Speak service.

no, I didn’t. I’m thinking it’s just a bug that’ll be patched in an upcoming release since this is really the beginning of Voice Assistant…

For me its just working using the link above that nickrout posted but i would guess its only working for this device at the moment.

Hi all,

I just updated to ESPHome 2023.5 and 2023.5.1 and it seems my echo is not working very well now, like when it talks back with a long reply, it studders and pops and does not finish the sentence. I have 3 of them and they all do the same now.

Also, the volume slider and media_player entity is gone, but I suspect it has something to with the new speaker ability, but not sure.

Anyone else experience this on update over the last couple of days? Was working very well on ESPHome 2023.4.4

issue reported here.

I updated mine recently and it started speaking back but I think the pipeline for TTS is broken because it’s responses are non sequitur. For example, I’ll say: “Turn off the office lights” and it will turn green, then actually turn off the office lights (!) but then in response, it’ll say: “Garage door is now open” which comes from another one of my automations but it is not related to the office lights! I’ve heard it say other things in response too.

Anyone know how to fix the TTS pipeline? I saw that there’s a clear cache option and I tried executing that but it didn’t seem to make a difference.

Yes! I see this broken pipiline too. Sometimes it does not work at all, goes blue then immediately red.

If you think your issue is related to the issue that I posted do you mind adding some of your observations from this? Whatever happened, it was working perfectly before 2023.5.

1 Like

Mine would not even update to the latest version… it showed me an error and stopped the update…

yes. I had the exact same problem. To solve this, place thins in your ESPHome yaml instead and it will work

packages:
  voice_assistant: 
    url: https://github.com/esphome/firmware
    file: voice-assistant/m5stack-atom-echo.yaml
    ref: main
    refresh: 0s

However, it was upgrading that has made all 3 of my Echo’s kinds useless as the responses are so choppy and incoherent. Not sure what here.

If you do update, please let me know if you experience similar issues, especially on long sentences like from OpenAI responses. thanks

To be honest this gives me other errors. But maybe because i dont where to exactly place it :slight_smile:

I’ve moved all my “speaking” automations back to the tts.google_say service for now because most of the time the tts.say+tts.piper voice would either not say anything or tell me something completely unrelated from the stale tts messaging pipeline.

I’ll watch the release notes for a fix on the pipeline to try it again. :confused:

If you go to ESPHome, press edit on echo and remove these lines

packages:
  esphome.bluetooth-proxy: github://esphome/bluetooth-proxies/esp32-generic.yaml@main

and replace with the block of text above.

Press save, and install.

Will try that later on … thanks. But i think latest esphome fix might be adressing that as well. Came out just minutes ago i guess :slight_smile:

That does not work for as i dont have these lines. What i have is:

dashboard_import:
  package_import_url: github://esphome/firmware/voice-assistant/m5stack-atom-echo.yaml@main

And if replace that one it shows me errors.

And the latest update of esphome btw. did not fix it as well…

What errors?