strong hearth Feb 22, 2024, 12:56 PM

#

this is the log for when i'm not commanding anything
[13:48:22][D][media_player:059]: 'ESPHome I2S Media Player' - Setting [13:48:22][D][media_player:066]: Media URL: http://192.168.1.239:8123/media/local/audio/activatedSound.mp3?authSig=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJpc3MiOiI3ZWI1OGIzZTQxZWM0M2YzYTM1ZDhkOGMxZGM1ODMyMiIsInBhdGgiOiIvbWVkaWEvbG9jYWwvYXVkaW8vYWN0aXZhdGVkU291bmQubXAzIiwicGFyYW1zIjpbXSwiaWF0IjoxNzA4NjA2MTAyLCJleHAiOjE3MDg2OTI1MDJ9.J6pJSNA5L0bZrot6JpiNDpL3556Id5El2dhLItpKHXc [13:48:37][D][voice_assistant:521]: Event Type: 12 [13:48:37][D][voice_assistant:679]: STT by VAD end [13:48:37][D][voice_assistant:414]: State changed from STREAMING_MICROPHONE to STOP_MICROPHONE [13:48:37][D][voice_assistant:420]: Desired state set to AWAITING_RESPONSE [13:48:37][D][voice_assistant:414]: State changed from STOP_MICROPHONE to STOPPING_MICROPHONE [13:48:37][D][voice_assistant:521]: Event Type: 0 [13:48:37][E][voice_assistant:651]: Error: stt-no-text-recognized - No text recognized [13:48:37][D][voice_assistant:514]: Signaling stop... [13:48:37][D][voice_assistant:414]: State changed from STOPPING_MICROPHONE to STOP_MICROPHONE [13:48:37][D][voice_assistant:420]: Desired state set to IDLE [13:48:37][D][voice_assistant:414]: State changed from STOP_MICROPHONE to IDLE [13:48:37][D][main:226]: Playback started! [13:48:37][W][component:214]: Component i2s_audio.media_player took a long time for an operation (0.56 s). [13:48:37][W][component:215]: Components should block for at most 20-30ms. [13:48:37][D][switch:016]: 'Use wake word' Turning OFF. [13:48:37][D][switch:055]: 'Use wake word': Sending state OFF [13:48:37][D][switch:012]: 'Use wake word' Turning ON. [13:48:37][D][switch:055]: 'Use wake word': Sending state ON [13:48:37][D][voice_assistant:414]: State changed from IDLE to START_PIPELINE

blissful sluice Feb 22, 2024, 12:57 PM

#

can you post your config

strong hearth Feb 22, 2024, 12:58 PM

#

`media_player:

platform: i2s_audio
name: ESPHome I2S Media Player
dac_type: external
i2s_dout_pin:
number: GPIO27 #DIN
allow_other_uses: true
mode: mono

on_play:

- delay: 700ms
- logger.log: "Playback started!"
- voice_assistant.stop:
on_idle:
- logger.log: "Playback finished!"
- voice_assistant.start_continuous:

output:

platform: gpio
pin:
number: GPIO27
allow_other_uses: true
id: set_low_speaker

voice_assistant:
microphone: mic_i2s
id: va
noise_suppression_level: 4
auto_gain: 31dBFS
volume_multiplier: 8.0
use_wake_word: false
speaker: my_speaker

on_wake_word_detected:
- homeassistant.service:
service: media_player.play_media
data:
entity_id: media_player.jarvis_esphome_i2s_media_player
media_content_id: >-
http://192.168.1.239:8123/media/local/audio/activatedSound.mp3
media_content_type: music
- switch.turn_on:
id: led_light
on_listening:
- switch.turn_on:
id: led_light

on_error:

if:
condition:
switch.is_on: use_wake_word
then:
- switch.turn_off: use_wake_word
- switch.turn_on: use_wake_word

on_client_connected:
- if:
condition:
switch.is_on: use_wake_word
then:
- voice_assistant.start_continuous:

on_client_disconnected:
- if:
condition:
switch.is_on: use_wake_word
then:
- voice_assistant.stop:
on_end:
- switch.turn_off:
id: led_light`

#

Do you need the full config?

blissful sluice Feb 22, 2024, 1:25 PM

#

that should be enough , i'll have a look through it in a short while 👍

#

what is the output for on gpio27? id: set_low_speaker ?

strong hearth Feb 22, 2024, 2:01 PM

#

set_low_speaker yes

strong hearth Feb 22, 2024, 2:24 PM

#

This is for fixing a buzz that happened on the speakers when powering on the device

blissful sluice Feb 22, 2024, 3:51 PM

#

try formatting the media url like below , if the file is in your media folder, and also move the block to 'on_listening' rather than 'on_wake_word_detected' and see what that does.

    - homeassistant.service:
        service: media_player.play_media
        data:
          media_content_id: media-source://media_source/local/vad.mp3
          media_content_type: music
          entity_id: media_player.tree_media

strong hearth Feb 22, 2024, 4:13 PM

#

thanks for the tip. Gonna try that tonight

blissful sluice Feb 22, 2024, 4:28 PM

#

I haven't physically set one up for a while to test as I only output to other media players and don't use one on the devices personally. however there are a few people trying to do the same, so I will put a test rig together later, makes it easier to advise , i've got the same setup 🙂

strong hearth Feb 22, 2024, 5:36 PM

#

This setup does not work 😦
`voice_assistant:
microphone: mic_i2s
id: va
noise_suppression_level: 4
auto_gain: 31dBFS
volume_multiplier: 8.0
use_wake_word: false
speaker: my_speaker

on_wake_word_detected:
- switch.turn_on:
id: led_light
on_listening:
- switch.turn_on:
id: led_light
- homeassistant.service:
service: media_player.play_media
data:
entity_id: media_player.jarvis_esphome_i2s_media_player
media_content_id: >-
media-source://media_source/local/audio/activatedSound.mp3
media_content_type: music`

and for some reason I can't see the logs via de wireless option
Error connecting to [AddrInfo(family=<AddressFamily.AF_INET: 2>, type=<SocketKind.SOCK_STREAM: 1>, proto=6, sockaddr=IPv4Sockaddr(address='192.168.1.140', port=6053))]: [Errno 113]

However I have set up the manual_ip option:
manual_ip: static_ip: 192.168.1.140 subnet: 255.255.255.0 gateway: 192.168.1.254

and the logs are filled with "Speaker buffer full, trying again next loop" error messages.

strong hearth Feb 23, 2024, 9:51 AM

#

Do you have any suggestions on how I could fix this? 🙂

blissful sluice Feb 23, 2024, 11:06 AM

#

hi, I have been having a look into this earlier this morning and got side tracked, I can get it to play the sound on wakeword, no problem and then it wont pickup the command. I know previously I have had it working but I can't find the config... so i'm having to figure it out again. Are you unable to get any sound at all from the device? even responses?

strong hearth Feb 23, 2024, 11:43 AM

#

It does recognize the wake word, led's turn on but still no sound, in the log it doesn't show that it is executing the service call anymore. previously I could see that in the log. And the audio response when the action is finished is very stutterly

#

It also doesn't work when I turn use wake word off to turn the pipeline off, and then do the service call via HA itself. No sound at all. But now it does show something in the logs
[D][media_player:059]: 'ESPHome I2S Media Player' - Setting [D][media_player:066]: Media URL: http://192.168.1.239:8123/media/local/audio/activatedSound.mp3?authSig=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJpc3MiOiJlZDhhMjdhY2ZhOGQ0MzM1OTM3ZTkwNGMxNmJhZTg1ZSIsInBhdGgiOiIvbWVkaWEvbG9jYWwvYXVkaW8vYWN0aXZhdGVkU291bmQubXAzIiwicGFyYW1zIjpbXSwiaWF0IjoxNzA4Njg4ODYwLCJleHAiOjE3MDg3NzUyNjB9.RhCvOVZLBURYIGpFpjGA7DOKe940xCJLjicPAj-pon4

#

well I need to leave soon, so it has no priority because I can't test anything.

blissful sluice Feb 23, 2024, 7:07 PM

#

I think you need to strip your config back to basics first , in order to rectify the overall sound issue, then start to build back up once the sound plays normally as it should. Once you have the basic config with audio working. then can look at adding the service calls etc.

strong hearth Feb 23, 2024, 7:09 PM

#

I´m uploading a new version without the service call to see if that fixes the sound issues

blissful sluice Feb 23, 2024, 7:09 PM

#

by basic config , just have wifi: ota: api: logger: 12s_audio: microphone : media_player : and a basic VA with no automations

blissful sluice Feb 23, 2024, 7:10 PM

#

strong hearth I´m uploading a new version without the service call to see if that fixes the so...

ok 👍 , what board, dac/amp is it ?

strong hearth Feb 23, 2024, 7:13 PM

#

The sound is fixed again, so it does work better without the service call at the on_listening block. But the logs are still filled with 'Speaker buffer full, trying again next loop' error messages.

I'm using a esp32dev board with MAX98357A dac and INMP441 mic

strong hearth Feb 23, 2024, 7:30 PM

#

Alright, so when I switch of the Use wake word switch, then I can play the audio file via the UI but not via a service call 🤔

blissful sluice Feb 23, 2024, 7:31 PM

#

a way I have also done it in the past is to use an automation in HA. Let me see if i can find that

strong hearth Feb 23, 2024, 7:33 PM

#

That's what I thought of as well. But would that be fast enough?

blissful sluice Feb 23, 2024, 7:35 PM

#

i think it worked ok iirc - so trigger = assist in progress then actions, switch ww off , wait for time = 200ms, then media play, then switch ww back on. worth a try and can adjust on the fly ... without having to keep re-compile the device. not ideal .. but a workaround without upsetting audio on the device

strong hearth Feb 23, 2024, 7:35 PM

#

ah right yes. Gonna try that now 🙂

blissful sluice Feb 23, 2024, 7:42 PM

#

good luck

#

I am still looking into getting it to trigger from the device side, I just have a couple of other projects on at the moment that are taking up a bit of time, so i start something then whilst waiting for something to upload or whatever .. i then start looking at another thing and then get deep into that so apologise if i'm a little slow at times getting back with ideas and info 😁

strong hearth Feb 23, 2024, 7:46 PM

#

haha thats no problem! I get it, I'm doing like 5 things at same time as well lol

#

I'm just glad that someone is willing to figure this out with me.

blissful sluice Feb 23, 2024, 7:50 PM

#

it is something that is asked for a lot, so i don't know why i haven't got a working answer lol.

strong hearth Feb 23, 2024, 7:54 PM

#

But we are getting somewhere! with the new automation it is starting to work! 🙂 When I trigger the wake word, the led turn on, then after i think 200ms they turn off and then the audio file plays. But it wont listen to the command

blissful sluice Feb 23, 2024, 7:57 PM

#

can you have the esphome logs open for the device to watch what happens, you should be able to see everything as it happens

strong hearth Feb 23, 2024, 8:18 PM

#

Yes I can see that.
Event Type: 10 [21:17:43][D][voice_assistant:530]: Wake word detected [21:17:43][D][switch:012]: 'Eyes' Turning ON. [21:17:43][D][switch:055]: 'Eyes': Sending state ON [21:17:43][D][voice_assistant:521]: Event Type: 3 [21:17:43][D][voice_assistant:535]: STT started [21:17:43][D][switch:012]: 'Eyes' Turning ON. [21:17:43][D][switch:016]: 'Use wake word' Turning OFF. [21:17:43][D][switch:055]: 'Use wake word': Sending state OFF [21:17:43][D][voice_assistant:514]: Signaling stop... [21:17:43][D][voice_assistant:414]: State changed from STREAMING_MICROPHONE to STOP_MICROPHONE [21:17:43][D][voice_assistant:420]: Desired state set to IDLE [21:17:43][D][voice_assistant:414]: State changed from STOP_MICROPHONE to STOPPING_MICROPHONE [21:17:43][D][voice_assistant:414]: State changed from STOPPING_MICROPHONE to IDLE [21:17:44][D][voice_assistant:521]: Event Type: 0 [21:17:44][E][voice_assistant:651]: Error: stt-no-text-recognized - No text recognized [21:17:44][D][voice_assistant:521]: Event Type: 2 [21:17:44][D][voice_assistant:611]: Assist Pipeline ended [21:17:44][D][switch:016]: 'Eyes' Turning OFF. [21:17:44][D][switch:055]: 'Eyes': Sending state OFF [21:17:44][D][media_player:059]: 'ESPHome I2S Media Player' - Setting [21:17:44][D][media_player:066]: Media URL: http://192.168.1.239:8123/media/local/audio/activatedSound.mp3? [21:17:44][W][component:214]: Component i2s_audio.media_player took a long time for an operation (0.54 s). [21:17:44][W][component:215]: Components should block for at most 20-30ms. [21:17:44][D][switch:012]: 'Use wake word' Turning ON. [21:17:44][D][switch:055]: 'Use wake word': Sending state ON

#

This is what happening in short

wake word said
wake word detected
LED turn on for like 200ms
LED turn off
audio file starts playing

blissful sluice Feb 23, 2024, 8:32 PM

#

can you post the config for your switch: component

strong hearth Feb 23, 2024, 8:34 PM

#

`switch:

platform: template
name: Use wake word
id: use_wake_word
optimistic: true
restore_mode: RESTORE_DEFAULT_ON
entity_category: config
on_turn_on:
- lambda: id(va).set_use_wake_word(true);
- if:
  condition:
  not:
  - voice_assistant.is_running
  then:
  - voice_assistant.start_continuous
on_turn_off:
- voice_assistant.stop
- lambda: id(va).set_use_wake_word(false);`

#

I have come a little further now. I right now I have the same behaviour but only with the config from the yaml. No automations. The sound does play finally. But only after it ended to listen for the command and the led turn off

blissful sluice Feb 23, 2024, 8:38 PM

#

i think something must have recently changes in how VA handles being stopped and started, because this [21:17:44][E][voice_assistant:651]: Error: stt-no-text-recognized - No text recognized i'm sure didn't used to happen, and the stream just started as soon as voice_assistant: was running, without the need for a wakeword. Well I know that it does work , as I have just out a push to talk in one of my builds... if i use wakeword it works. but if i just press the 'button' i can say a command without ww. which really is what's happening here ... so it's a little odd that it's giving that error

blissful sluice Feb 23, 2024, 8:40 PM

#

strong hearth I have come a little further now. I right now I have the same behaviour but only...

yes, so at the point the mic is off and the media can play. and all we want to do is turn mic off to let audio play then turn back on to give the command.! sounds simple

strong hearth Feb 23, 2024, 8:41 PM

#

Yes that is what we want. But it doesn't do that 😦

blissful sluice Feb 23, 2024, 8:42 PM

#

if only things worked as well in practice as they do in theory. and the stupid thing is, i've done it before. but because I don't actually do it that way myself... it's been lost in configs and my mind lol

strong hearth Feb 23, 2024, 8:42 PM

#

has this error 'Component i2s_audio.media_player took a long time for an operation (0.54 s).' something to do with that as well or no?

#

You do it via automations? or only push to talk?

blissful sluice Feb 23, 2024, 8:43 PM

#

no thats fine, thats just a warning , that only shows because they changed the logging level in ESPHome a few months ago, previously the errors were still there but supressed

strong hearth Feb 23, 2024, 8:43 PM

#

ah right

#

and the annoying thing is that other people have figured it out and got it working perfectly like this
https://community.home-assistant.io/t/r2d2-satellite/690701/8
so it is possible, and I wont stop trying until if have it working xD

blissful sluice Feb 23, 2024, 8:46 PM

#

strong hearth You do it via automations? or only push to talk?

I don't use any devices with mic and speaker. i have them on separate devices, i so i can have mics in different areas and speakers just in central areas, if that makes sense ( sounds like i live in a castle lol ) i don't need audio feedback everywhere

blissful sluice Feb 23, 2024, 8:47 PM

#

strong hearth and the annoying thing is that other people have figured it out and got it worki...

ah yes... on the wyoming satellite it just works! i've got one of those and it's great lol

strong hearth Feb 23, 2024, 8:47 PM

#

Oh really. I wanted to start that project once I have this one working. This will be the main "Jarvis" in the living room

#

3d printed helmet and all

#

and I was thinking of making some little r2d2 devices as satellites

blissful sluice Feb 23, 2024, 8:49 PM

#

have you got room in it to put another esp ? as that is a solution and works ! and at the cost of an esp doesnt brake the bank. I've built a lamp that I finished yesterday and I put 2 in that 1 , (well 3 in total lol)

#

https://discord.com/channels/429907082951524364/1210369185612562512

strong hearth Feb 23, 2024, 8:51 PM

#

Oh wow nice! also with touch trigger. Thats a nice addition!

blissful sluice Feb 23, 2024, 8:53 PM

#

yes. i was quite suprised how well the touch works, and it also works through the painted outer. and all that uses is a piece of wire on a gpio . using the esp32_touch: component

strong hearth Feb 23, 2024, 8:54 PM

#

oh thats a very clever idea yes!

blissful sluice Feb 23, 2024, 8:55 PM

#

but with that. I have the mic on 1 esp32 and the media on another. and that way it's no problem calling the media to play whilst the va is running

#

i did it that way to share the load , with using onboard wakeword, an oled, a mmwave sensor and temp sensor etc

strong hearth Feb 23, 2024, 8:56 PM

#

oh right, the only tutorial that I have watched uses a raspberry pi zero 2 w and a mic hat? that is what seperates them?

blissful sluice Feb 23, 2024, 8:58 PM

#

the pizerow uses the mic hat which is basically a soundcard in it's own right. so it can handle the audio side of things much better

#

and it isn't limited by the esp voice_assistant limitations of only being able to do either mic or speaker at a time and not both together

strong hearth Feb 23, 2024, 9:06 PM

#

ohhh thats why that is working so flawelessly! I am for sure gonna make that after this project

#

btw it just good to know that I have changed some thing with the media play service call.

I came to the conclusion that I was trying to play on the wrong entity_id xD
and I changed media_content_id: to -> media-source://media_source/local/audio/{audio_file}
and I changed the media_content_type: from music -> audio/mpeg

#

but it should be doable what we are trying to do right?

blissful sluice Feb 23, 2024, 9:14 PM

#

unless something has changed in the VA code that's stopping it. right , I think ive just caught up with everything. so i can plug my test device in and have another play.

strong hearth Feb 23, 2024, 9:15 PM

#

ah nice 🙂

blissful sluice Feb 23, 2024, 9:19 PM

#

unfortunatley the only one with media player to hand is an atom echo! so I can't hear it very well anyway 🙂

strong hearth Feb 23, 2024, 9:20 PM

#

you mean the tts? Whats important that it works right xD optimalization comes later

blissful sluice Feb 23, 2024, 9:22 PM

#

totally! just going to add a switch to the config and then try from HA

strong hearth Feb 23, 2024, 9:24 PM

#

what are you trying to make right now? also a voice assistant with esp?

blissful sluice Feb 23, 2024, 9:32 PM

#

i haven't got any projects on the go at the moment, just need to finish the final bits on that lamp. but i'm just testing stuff now and trying to figure out how to get stuff to work 🙂

strong hearth Feb 23, 2024, 9:32 PM

#

ah oke cool

blissful sluice Feb 23, 2024, 9:42 PM

#

ooops totally broke the config lol , need to did another out to fix it

strong hearth Feb 23, 2024, 9:52 PM

#

lol happens a lot sometimes, but once it starts working it all fine

blissful sluice Feb 23, 2024, 9:59 PM

#

just compiling

#

hmm same error Error: stt-no-text-recognized - No text recognized

strong hearth Feb 23, 2024, 10:15 PM

#

yea it's so annoying, im trying to make it work as well. But still the same results as before

blissful sluice Feb 23, 2024, 10:18 PM

#

that didn't used to come up i'm sure hmm

strong hearth Feb 23, 2024, 10:24 PM

#

I think whats going wrong as well is that once I do the voice_assistant.start_continuous it cancels the listening phase or something? but the weird thing is that the audio file only plays after the listening phase is done

blissful sluice Feb 23, 2024, 10:33 PM

#

I need to take a break for while, i'm getting brain fog ... been at it since 3am i think i need some food to refuel.

strong hearth Feb 23, 2024, 10:46 PM

#

good idea. I need to get some sleep as well in a bit. its 11:45 pm where i am

blissful sluice Feb 23, 2024, 10:49 PM

#

yeah 22.49 here now

#

i'll pick it up again tomorrow with a fresh set of eyes 🙂

strong hearth Feb 23, 2024, 10:53 PM

#

me too, have a good night

blissful sluice Feb 23, 2024, 10:54 PM

#

cheers , you too catcha later

blissful sluice Feb 24, 2024, 5:55 AM

#

well some people have been sleeping 🤪 i might have sorted it!!!

strong hearth Feb 24, 2024, 10:07 AM

#

really?! is it working for you?

blissful sluice Feb 24, 2024, 10:07 AM

#

so far so good!

#

the occaisional error, but it carries on working and doesn't cause it to fail

strong hearth Feb 24, 2024, 10:09 AM

#

How did you manage to fix it? 🙂

blissful sluice Feb 24, 2024, 10:10 AM

#

if you transfer 10,000000 into my bitcoin account i will reveal the secrets 🤣

#

ok it was really simple in the end!

strong hearth Feb 24, 2024, 10:10 AM

#

you mean you fixed the wake word detected sound right? 🙂

blissful sluice Feb 24, 2024, 10:11 AM

#

yes, you say hey wakeword - and it plays a sound and you say .turn on the light, and it does it!

strong hearth Feb 24, 2024, 10:11 AM

#

yessss

blissful sluice Feb 24, 2024, 10:12 AM

#

would you like my code

#

you'll kick yourself whne you see it!

strong hearth Feb 24, 2024, 10:12 AM

#

Please show me, im very curious right now xD

blissful sluice Feb 24, 2024, 10:12 AM

#

voice_assistant:
  id: va
  microphone: echo_microphone
  media_player: echo_speaker
  noise_suppression_level: 2
  auto_gain: 31dBFS
  volume_multiplier: 4.0
  use_wake_word: true
  on_wake_word_detected:
    - voice_assistant.stop
    - lambda: id(va).set_use_wake_word(false);
    - delay: 100ms
    - media_player.play_media: http://192.168.1.8:8123/local/vad.mp3
    - delay: 100ms
    - voice_assistant.start_continuous
  on_tts_end:
        then:
        - homeassistant.service:
            service: media_player.play_media
            data:
              entity_id: media_player.m5media
              media_content_id: !lambda 'return x;'
              media_content_type: music
              announce: "true"
        - delay: 50ms
        - switch.turn_on: use_wake_word

strong hearth Feb 24, 2024, 10:12 AM

#

is it the announce true??

blissful sluice Feb 24, 2024, 10:13 AM

#

its just the order that the actions are listed under on_wake_word_detected:

strong hearth Feb 24, 2024, 10:13 AM

#

oh no thats the normal response from the action

#

ohh right. But I kinda had that already why wouldn't that work xD im gonna test this right now !

blissful sluice Feb 24, 2024, 10:15 AM

#

and also after on_tts_end add the delay and the switch.turn on

#

thes last 2 entries in the code

strong hearth Feb 24, 2024, 10:15 AM

#

on_tts_end is not the same as on_end?

blissful sluice Feb 24, 2024, 10:16 AM

#

no

strong hearth Feb 24, 2024, 10:17 AM

#

and at the on wake word detected block you platy the sound via the url, but no entity specified?

blissful sluice Feb 24, 2024, 10:17 AM

#

correct

#

it uses the media player configured on the device, which is how we wanted it originally , but tried with service calls ... but this 100% works for me

strong hearth Feb 24, 2024, 10:18 AM

#

compiling the code now 🙂

blissful sluice Feb 24, 2024, 10:18 AM

#

🤞

strong hearth Feb 24, 2024, 10:18 AM

#

I hope it works!

#

omg it took so long that I had a timeout: Error receiving acknowledge version: timed out

blissful sluice Feb 24, 2024, 10:25 AM

#

🤦‍♂️

#

if it doesn't work, i'm hiding

strong hearth Feb 24, 2024, 10:26 AM

#

uploading directly to the esp right now instead of wirelessly

#

btw do you know if there is a way to rename the COM port devices in device manager? I have multiple connected so sometimes it difficult to find the right device

#

its not working yet 😦
the logs are spammed with
[11:29:32][E][voice_assistant:651]: Error: no_wake_word - No wake word detected [11:29:32][D][voice_assistant:514]: Signaling stop... [11:29:32][D][voice_assistant:414]: State changed from STARTING_PIPELINE to STOP_MICROPHONE [11:29:32][D][voice_assistant:420]: Desired state set to IDLE [11:29:32][D][voice_assistant:414]: State changed from STOP_MICROPHONE to IDLE [11:29:32][D][voice_assistant:521]: Event Type: 2 [11:29:32][D][voice_assistant:611]: Assist Pipeline ended [11:29:32][D][voice_assistant:414]: State changed from IDLE to START_PIPELINE [11:29:32][D][voice_assistant:420]: Desired state set to START_MICROPHONE [11:29:32][D][voice_assistant:514]: Signaling stop... [11:29:32][D][voice_assistant:118]: microphone not running [11:29:32][D][voice_assistant:202]: Requesting start... [11:29:32][D][voice_assistant:414]: State changed from START_PIPELINE to STARTING_PIPELINE [11:29:32][D][voice_assistant:521]: Event Type: 1 [11:29:32][D][voice_assistant:524]: Assist Pipeline running [11:29:32][D][voice_assistant:118]: microphone not running [11:29:32][D][voice_assistant:521]: Event Type: 9 [11:29:32][D][voice_assistant:118]: microphone not running [11:29:32][D][voice_assistant:521]: Event Type: 0
Over and over again

blissful sluice Feb 24, 2024, 10:31 AM

#

unplug and replug

#

it's gone very quiet over there ... is that a good or a bad thing ?

strong hearth Feb 24, 2024, 10:49 AM

#

kinda bad thing 😦 it's not really working anymore. Even sound is not playing 🤔

#

this is the config right now
`voice_assistant:
microphone: mic_i2s
id: va
noise_suppression_level: 4
auto_gain: 31dBFS
volume_multiplier: 8.0
use_wake_word: false
speaker: my_speaker

on_wake_word_detected:
- switch.turn_on:
id: led_light
- voice_assistant.stop
- lambda: id(va).set_use_wake_word(false);
- delay: 100ms
- media_player.play_media: media-source://media_source/local/audio/activatedSound.mp3
- delay: 100ms
- voice_assistant.start_continuous

on_tts_end:
then:
- homeassistant.service:
service: media_player.play_media
data:
entity_id: media_player.ultimate_jarvis_esphome_i2s_media_player
media_content_id: !lambda 'return x;'
media_content_type: music
announce: "true"
- delay: 50ms
- switch.turn_on: use_wake_word

on_listening:
- switch.turn_on:
id: led_light

on_error:

if:
condition:
switch.is_on: use_wake_word
then:
- switch.turn_off: use_wake_word
- switch.turn_on: use_wake_word

on_client_connected:
- if:
condition:
switch.is_on: use_wake_word
then:
- voice_assistant.start_continuous:

on_client_disconnected:
- if:
condition:
switch.is_on: use_wake_word
then:
- voice_assistant.stop:
on_end:
- switch.turn_off:
id: led_light`

#

I can try to play the sound via ip url

#

and now I have this issue again...
Can't connect to ESPHome API for ultimate-jarvis @ 192.168.1.140: Error connecting to [AddrInfo(family=<AddressFamily.AF_INET: 2>, type=<SocketKind.SOCK_STREAM: 1>, proto=6, sockaddr=IPv4Sockaddr(address='192.168.1.140', port=6053))]: [Errno 113] Connect call failed ('192.168.1.140', 6053) (SocketAPIError)

#

ill be back soon! just gotta do something with family

blissful sluice Feb 24, 2024, 10:53 AM

#

ok no worries, i'll be about and we WILL get it working!

strong hearth Feb 24, 2024, 12:22 PM

#

alright i'm back!

#

Lets get this thing working 🙂

blissful sluice Feb 24, 2024, 12:23 PM

#

ok boss

#

first thing i notices is that you have use_wake_word: false in the voice_assistant section

#

needs to be true

#

but well need to get you connected to the device again , can you paste the top part of your config, before the voice stuff you posted earler

strong hearth Feb 24, 2024, 12:26 PM

#

blissful sluice but well need to get you connected to the device again , can you paste the top p...

I replugged the device in, seems to be working again now.

blissful sluice Feb 24, 2024, 12:26 PM

#

ok cool

strong hearth Feb 24, 2024, 12:27 PM

#

I changed the use_wake_word to true

blissful sluice Feb 24, 2024, 12:28 PM

#

nothing else stands out from what i can see, so upload and try again, after flashing unplugit, give it a few seconds and plug it back in again

strong hearth Feb 24, 2024, 12:29 PM

#

still no working sound.. also when I give it the command I get a notification in HA
Login attempt or request with invalid authentication from 192.168.1.140 (192.168.1.140). See the log for details.

blissful sluice Feb 24, 2024, 12:30 PM

#

ok , so it looks like it needs reconfiguring in HA - is there an api key in your config ?

strong hearth Feb 24, 2024, 12:31 PM

#

yes there is

#

I do also have this in the config
manual_ip: static_ip: 192.168.1.140 subnet: 255.255.255.0 gateway: 192.168.1.254

but that shouldn't give problems right?

blissful sluice Feb 24, 2024, 12:33 PM

#

ok you will need that api - what you need to do is delete the device from the integration but not the addon. then from the addon view the logs. this should then trigger HA to discover the device again. Having the static in there shouldn't be an issue. do you have a static reserved for it in the router?

strong hearth Feb 24, 2024, 12:35 PM

#

blissful sluice ok you will need that api - what you need to do is delete the device from the i...

No i don't have it set as static ip in my router. I will add that as well. And rediscover the device in HA.

blissful sluice Feb 24, 2024, 12:37 PM

#

yes, if you don't have it reserved in the router, then it can give that address to another device, then you run into problems.

#

once you have added it back to HA if you look down the list of devices in the ESPHome integration, there will be a configure button next to each device, click that and then tick the box to allow service calls

strong hearth Feb 24, 2024, 12:51 PM

#

still the same login attempt error message 🤔 and what I have noticed is that the device in HA only has the sensor 'Assist in progress' and nothing else. No media player no eyes switch

#

oh wait now there in the device in HA

blissful sluice Feb 24, 2024, 12:58 PM

#

it will take a moment for the device to update the entities whne you first add it... how you getting on

strong hearth Feb 24, 2024, 1:02 PM

#

this is the voice assistant config block right now
`voice_assistant:
microphone: mic_i2s
id: va
noise_suppression_level: 4
auto_gain: 31dBFS
volume_multiplier: 8.0
use_wake_word: true
speaker: my_speaker

on_wake_word_detected:
- switch.turn_on:
id: led_light
- voice_assistant.stop
- lambda: id(va).set_use_wake_word(false);
- delay: 100ms
- media_player.play_media: http://192.168.1.239:8123/media/local/audio/activatedSound.mp3
- delay: 100ms
- voice_assistant.start_continuous

on_tts_end:
then:
- homeassistant.service:
service: media_player.play_media
data:
entity_id: media_player.ultimate_jarvis_esphome_i2s_media_player
media_content_id: !lambda 'return x;'
media_content_type: music
announce: "true"
- delay: 50ms
- switch.turn_on: use_wake_word
- switch.turn_off:
id: led_light

on_error:

if:
condition:
switch.is_on: use_wake_word
then:
- switch.turn_off: use_wake_word
- switch.turn_on: use_wake_word

on_client_connected:
- if:
condition:
switch.is_on: use_wake_word
then:
- voice_assistant.start_continuous:

on_client_disconnected:
- if:
condition:
switch.is_on: use_wake_word
then:
- voice_assistant.stop:

on_end:
- switch.turn_off:
id: led_light`

What i have noticed is that when I turn off the wake word switch manually and then trigger the sound via the service call like we did before, then it works

#

So I think I need to try the service call again?

blissful sluice Feb 24, 2024, 1:05 PM

#

you can try it , but it seems strange

strong hearth Feb 24, 2024, 1:08 PM

#

the sound does work again. But still only after the led eyes turned off and the pipeline ended

#

just for a test I will turn off all the things related to the LED eyes

blissful sluice Feb 24, 2024, 1:10 PM

#

yes it's worth trying as minimal as possible

#

esphome:
  name: m5stack-atom-echo-889258
  friendly_name: M5Stack Atom Echo 889258

api:
  on_client_connected:
   - delay: 500ms
   - switch.turn_on: use_wake_word

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

esp32:
  board: m5stack-atom
  framework:
    type: arduino

logger:
ota:

button:
  - platform: restart
    id: restart_but
    name: Reboot Device

i2s_audio:
  i2s_lrclk_pin: GPIO33
  i2s_bclk_pin: GPIO19

microphone:
  - platform: i2s_audio
    id: echo_microphone

    i2s_din_pin: GPIO23
    adc_type: external
    pdm: true
    
media_player:
  - platform: i2s_audio
    name: m5media
    id: echo_speaker
    i2s_dout_pin: 22
    dac_type: external
    mode: mono

voice_assistant:
  id: va
  microphone: echo_microphone
  media_player: echo_speaker
  noise_suppression_level: 2
  auto_gain: 31dBFS
  volume_multiplier: 4.0
  use_wake_word: true
  on_wake_word_detected:
    - voice_assistant.stop
    - lambda: id(va).set_use_wake_word(false);
    - delay: 100ms
    - media_player.play_media: http://192.168.1.8:8123/local/vad.mp3
    - delay: 100ms
    - voice_assistant.start_continuous
  on_tts_end:
        then:
        - homeassistant.service:
            service: media_player.play_media
            data:
              entity_id: media_player.m5media
              media_content_id: !lambda 'return x;'
              media_content_type: music
              announce: "true"
        - delay: 50ms
        - switch.turn_on: use_wake_word
switch:
  - platform: template
    name: Use wake word
    id: use_wake_word
    optimistic: true
    entity_category: config
    on_turn_on:
      - lambda: id(va).set_use_wake_word(true);
      - delay: 200ms
      - voice_assistant.start_continuous
    on_turn_off:
      - voice_assistant.stop
      - lambda: id(va).set_use_wake_word(false);

#

that is my complete yaml for the test device i used

strong hearth Feb 24, 2024, 1:24 PM

#

alright thanks, I have changed some things based on your config. Maybe it will work now.

blissful sluice Feb 24, 2024, 1:32 PM

#

heres hoping 😬

strong hearth Feb 24, 2024, 1:33 PM

#

you dont use LED in your assistant right?

blissful sluice Feb 24, 2024, 1:38 PM

#

thats just a test rig, I have only just got round to setting them up for myself to use. I just try different things to help people out

strong hearth Feb 24, 2024, 1:39 PM

#

ahh right

blissful sluice Feb 24, 2024, 1:40 PM

#

i've got quite a few setups for testing with different boards etc .. it's much easier to help if you can replicate the setup

strong hearth Feb 24, 2024, 1:44 PM

#

Im still getting these 'Error: stt-no-text-recognized - No text recognized' error messages..

#

this is my config now
`voice_assistant:
microphone: mic_i2s
id: va
noise_suppression_level: 4
auto_gain: 31dBFS
volume_multiplier: 8.0
use_wake_word: true

speaker: my_speaker

strong hearth · 2024-02-22T12:56:15.884Z

Sound effect at wake word detected not playing | Home Assistant | Page 1

media_player: jarvis_speaker

on_wake_word_detected:
# - switch.turn_on:
# id: led_light
- voice_assistant.stop
- lambda: id(va).set_use_wake_word(false);
- delay: 100ms
- homeassistant.service:
service: media_player.play_media
data:
entity_id: media_player.ultimate_jarvis_esphome_i2s_media_player
media_content_id: !lambda 'return "http://192.168.1.239:8123/media/local/audio/activatedSound.mp3";'
media_content_type: audio/mpeg
- delay: 100ms
- voice_assistant.start_continuous

on_tts_end:
then:
- homeassistant.service:
service: media_player.play_media
data:
entity_id: media_player.ultimate_jarvis_esphome_i2s_media_player
media_content_id: !lambda 'return x;'
media_content_type: music
announce: "true"
- delay: 50ms
- switch.turn_on: use_wake_word
# - switch.turn_off:
# id: led_light`

blissful sluice Feb 24, 2024, 1:47 PM

#

how long is the sound

#

as in time duration

#

try increasing the 2 x 100ms delays to 200ms

#

and is the sound a sound or a voice?

strong hearth Feb 24, 2024, 1:58 PM

#

Its a sound, windows player says its 0:00 secs😅

blissful sluice Feb 24, 2024, 1:59 PM

#

thats fast 🤣

#

or should i say short... i mean it plays and stops fast lol

#

you could to test with just copy and paste my whole config and just change the device name, network and pin numbers and name of mp3 to see if that works.

#

what are you using for tts and stt >

#

?

strong hearth Feb 24, 2024, 2:07 PM

#

blissful sluice you could to test with just copy and paste my whole config and just change the d...

I tried that, but that doesn't work. The audio file will only play with the service call

strong hearth Feb 24, 2024, 2:07 PM

#

blissful sluice what are you using for tts and stt >

home assistant cloud

#

I just changed the delay from 100ms to 200ms, lets see if that is better

blissful sluice Feb 24, 2024, 2:08 PM

#

same a sme , so that shouldnt cause an issue. it makes no sense that it will play from a service call , as it's calling what your'e doing internally on the device. where abouts in HA is the file? is it in media or just in a folder in www

#

if you put the url in a broswer does it play the file

strong hearth Feb 24, 2024, 2:11 PM

#

blissful sluice if you put the url in a broswer does it play the file

im getting a 401: unauthorized error

blissful sluice Feb 24, 2024, 2:12 PM

#

yeah it's the wrong path in that case

#

oh no .. sorry a 401 . hmm

strong hearth Feb 24, 2024, 2:12 PM

#

and the audio file is in media -> my media -> audio -> activatedSound.mp3

#

its giving me the same error message when I try to acces it via the external nabu casa url

#

so https

blissful sluice Feb 24, 2024, 2:14 PM

#

ive just put mine in the www folder so it accessible from /local

strong hearth Feb 24, 2024, 2:18 PM

#

hmm will try that for a change xD

#

hey! that works. It does play the sound when I acces that file via url

blissful sluice Feb 24, 2024, 2:20 PM

#

a step closer to success

strong hearth Feb 24, 2024, 2:21 PM

#

lets hope for some progress

#

trying it now with your config line

#

it wont recognize the wake word now 😦 and the logs are constantly filled with these logs
'[15:22:18][D][voice_assistant:521]: Event Type: 0
[15:22:18][E][voice_assistant:651]: Error: wake-stream-failed - Unexpected error during wake-word-detection
[15:22:18][D][voice_assistant:514]: Signaling stop...
[15:22:18][D][voice_assistant:414]: State changed from STREAMING_MICROPHONE to STOP_MICROPHONE
[15:22:18][D][voice_assistant:420]: Desired state set to IDLE
[15:22:18][D][voice_assistant:414]: State changed from STOP_MICROPHONE to STOPPING_MICROPHONE
[15:22:18][D][voice_assistant:521]: Event Type: 2
[15:22:18][D][voice_assistant:611]: Assist Pipeline ended
[15:22:18][D][voice_assistant:414]: State changed from STOPPING_MICROPHONE to IDLE
[15:22:18][D][voice_assistant:514]: Signaling stop...
[15:22:18][D][voice_assistant:414]: State changed from IDLE to START_PIPELINE
[15:22:18][D][voice_assistant:420]: Desired state set to START_MICROPHONE
[15:22:18][D][voice_assistant:521]: Event Type: 1
[15:22:18][D][voice_assistant:524]: Assist Pipeline running
[15:22:18][D][voice_assistant:118]: microphone not running
[15:22:18][D][voice_assistant:202]: Requesting start...
[15:22:18][D][voice_assistant:414]: State changed from START_PIPELINE to STARTING_PIPELINE
[15:22:18][D][voice_assistant:521]: Event Type: 9'

#

adding the on_error block back now to see if that fixes the spam\

blissful sluice Feb 24, 2024, 2:25 PM

#

did you unplug after flashing

strong hearth Feb 24, 2024, 2:25 PM

#

no?

blissful sluice Feb 24, 2024, 2:25 PM

#

just unplug and replug if you get errors after flashing , best to have a full reboot

strong hearth Feb 24, 2024, 2:26 PM

#

ah right, thanks for the tip

blissful sluice Feb 24, 2024, 2:28 PM

#

i'm just pluging my test one back in to make sure i wasnt imagining it worked 😄

strong hearth Feb 24, 2024, 2:28 PM

#

something is going wrong when this error occurs:
Error: stt-no-text-recognized - No text recognized
because then the pipeline gets reset or something, and it wont listen for the command anymore

#

this error happens everytime I trigger the wake word 🤔 but not when its just waiting for the wake word to be detected

strong hearth Feb 24, 2024, 2:30 PM

#

blissful sluice i'm just pluging my test one back in to make sure i wasnt imagining it worked 😄

lol

blissful sluice Feb 24, 2024, 2:33 PM

#

strong hearth this error happens everytime I trigger the wake word 🤔 but not when its just w...

its because its expecting a sound, mines not working now 🤣 thats ridiculous, it keeps just leaving the mic open so i dont need to use wakeword

#

which is quite good really hahahah

#

just say turn off kitchen and it does without wake word 🤬

#

it was perfect this morning

strong hearth Feb 24, 2024, 2:35 PM

#

blissful sluice its because its expecting a sound, mines not working now 🤣 thats ridiculous, i...

ohh that sucks! It's so annoying that something will work one time and then it doesn't anymore

blissful sluice Feb 24, 2024, 2:37 PM

#

right.... i need a break from it , and then it will work! it will not beat me! man verses machine!

strong hearth Feb 24, 2024, 2:37 PM

#

hahaha we are not giving up indeed!

#

yoooo it worked for a moment! but then the whole device crashed xD

#

Guru Meditation Error: Core 1 panic'ed (LoadProhibited). Exception was unhandled.

#

wow and it worked for a second time as well. but still with the crash lol

#

So I guess this needs to be fixed by nabu casa

#

This error means the CPU was trying to load from an illegal address in memory.

#

This is what I read online

blissful sluice Feb 24, 2024, 2:44 PM

#

i'll have a look again shortly... need some food and sort a few things

strong hearth Feb 24, 2024, 2:45 PM

#

Thats okay, take your time. No worries

blissful sluice Feb 25, 2024, 7:57 AM

#

it works withoout any messing using an ESP32-S3 and micro_wake_word 🤦‍♂️

strong hearth Feb 25, 2024, 10:44 AM

#

what do you mean?

#

You got it working again?

blissful sluice Feb 25, 2024, 10:47 AM

#

i was just testing with a different setup. using local wake word (micro_wake_word) it works fine, but you need an esp32-s3 with psram it doesnt work on a normal esp32

strong hearth Feb 25, 2024, 10:48 AM

#

Yea im thinking of ordering a board as well soon. Just ordered a raspberry pi zero 2 w with mic HAT to make a satellite.

#

Does it work with the wake sound as well?

blissful sluice Feb 25, 2024, 10:49 AM

#

yes, the pi zero does 👍

#

everything is moving quite quickly with esp's now and a lot of new projects like voice, cameras, media etc need something with psram and the S3n16r8 fits the bill well

strong hearth Feb 25, 2024, 10:52 AM

#

Yes it does. very exciting weeks to see all the progress everyone has made with these projects

blissful sluice Feb 25, 2024, 10:53 AM

#

also esp-idf supported media player might arrive quite soon... but will have to wait and see what happens

strong hearth Feb 25, 2024, 10:53 AM

#

I reset my config back to when it was working with only LED on wake word detected. For now at least it works and I'll have a look at it soon again. Maybe someone else will figure it out and I can use that config for the sound issue

strong hearth Feb 25, 2024, 10:54 AM

#

blissful sluice also esp-idf supported media player might arrive quite soon... but will have to ...

Oh really? that could be quite interesting

#

I was looking for the board
https://nl.aliexpress.com/item/1005006427985041.html?spm=a2g0o.detail.pcDetailTopMoreOtherSeller.1.5e3a5942XXyuYr&gps-id=pcDetailTopMoreOtherSeller&scm=1007.40050.354490.0&scm_id=1007.40050.354490.0&scm-url=1007.40050.354490.0&pvid=b1657fcd-612a-4f5f-9307-ab618a4dc1e8&_t=gps-id:pcDetailTopMoreOtherSeller,scm-url:1007.40050.354490.0,pvid:b1657fcd-612a-4f5f-9307-ab618a4dc1e8,tpp_buckets:668%232846%238111%231996&pdp_npi=4%40dis!EUR!11.17!3.66!!!85.04!27.89!%40211b600517088582575765325e62a3!12000037132978617!rec!NL!845371626!&utparam-url=scene%3ApcDetailTopMoreOtherSeller|query_from%3A

but I don't know if this as enough pins

blissful sluice Feb 25, 2024, 10:55 AM

#

yes, with an s3 local wake word and esp-idf media , it is pretty much a match made in heaven... everything should play nicely together

strong hearth Feb 25, 2024, 10:55 AM

#

blissful sluice yes, with an s3 local wake word and esp-idf media , it is pretty much a match ma...

I hope so. Lets hope it will come out soon and we can play with it 🙂

blissful sluice Feb 25, 2024, 10:56 AM

#

i'm already testing it 😉

strong hearth Feb 25, 2024, 10:56 AM

#

Next project for me will be setting up the satellites

#

So it does already work with a normal s3 board then?

blissful sluice Feb 25, 2024, 10:58 AM

#

those boards i'll have a look at the specs, i've got loads of those c3 ones and they are great for mic only .and sensors... wled and relays etc. .. very low power too ..work really well. but no media support for the c3 . just need to make sure that it is an s3 with PSRAM it needs that additional ram

strong hearth Feb 25, 2024, 10:59 AM

#

It does have s3 version as well on the page. Its the second option

blissful sluice Feb 25, 2024, 11:01 AM

#

strong hearth So it does already work with a normal s3 board then?

just looking and it doesn't give any specs .. hmmm it needs to have psram else it's no good

strong hearth Feb 25, 2024, 11:02 AM

#

in the description when you expand it. You can see the specs of the s3 zero. There it does say it has 2MB of PSRAM

blissful sluice Feb 25, 2024, 11:02 AM

#

https://www.aliexpress.com/item/1005004637451669.html these are the full size boards i use a lot

blissful sluice Feb 25, 2024, 11:02 AM

#

strong hearth in the description when you expand it. You can see the specs of the s3 zero. The...

oh in that case i will order some

strong hearth Feb 25, 2024, 11:03 AM

#

blissful sluice https://www.aliexpress.com/item/1005004637451669.html these are the full size bo...

These are a better choice I think? How many pins do you need for a working setup? I don't have it at hand at,

#

atm*

blissful sluice Feb 25, 2024, 11:05 AM

#

for a voice setup will use 3 for mic 3 for dac and power

#

just ordered some of the minis i like the size of those boards

strong hearth Feb 25, 2024, 11:06 AM

#

blissful sluice just ordered some of the minis i like the size of those boards

haha nice

strong hearth Feb 25, 2024, 11:07 AM

#

blissful sluice for a voice setup will use 3 for mic 3 for dac and power

and you'll need more for speakers as well

blissful sluice Feb 25, 2024, 11:07 AM

#

i shouldn't really as i already have loads and already another 15 of the big boards on the way 🤣 but i can't resist

blissful sluice Feb 25, 2024, 11:07 AM

#

strong hearth and you'll need more for speakers as well

6 in total + led

strong hearth Feb 25, 2024, 11:08 AM

#

Yes I just counted for wires going into the board with the setup I'm using right now

blissful sluice Feb 25, 2024, 11:08 AM

#

they do have an onboard addressable led wchis is quite bright, so depending on project that can act as the wake , progress light etc

strong hearth Feb 25, 2024, 11:10 AM

#

I use those https://nl.aliexpress.com/item/4001313545899.html?spm=a2g0o.order_list.order_list_main.32.7be079d2pAc4lU&gatewayAdapt=glo2nld for led to act as wake light 🙂

#

just a simple positive and negative

#

Jarvis will take over the house soon xD

blissful sluice Feb 25, 2024, 11:12 AM

#

oh thats an easy but effective solution 🙂

strong hearth Feb 25, 2024, 11:13 AM

#

it does work pretty well yes 🙂

#

only thing that is left is this freaking wake word sound.. but maybe it can be done with the new s3 boards

blissful sluice Feb 25, 2024, 11:15 AM

#

100% works with S3 and local wakeword ... because the pipeline doesnt contstantly run and the wake word is completley seperate, so after wake word detected, you can run whatever you want befor opening the pipeline for voice

strong hearth Feb 25, 2024, 11:16 AM

#

ohhhh yes thats true! haven't thought about that. Yes then it should work 🙂 Im gonna order some boards now

#

do you recommend the bigger ones or the smaller?

#

I need to 4 gpio pins, v5, gnd and 3v3

blissful sluice Feb 25, 2024, 11:19 AM

#

they will both run it... personally if you have the space in your project i would go for the larger it has 16mb flash and 8mb ram... so if you want to add more complex stuff in the future then you have scope to expand

strong hearth Feb 25, 2024, 11:20 AM

#

I have the full iron man helmet to put stuff in, so space enough. I will order some larger boards then

blissful sluice Feb 25, 2024, 11:21 AM

#

oh cool ... lol , my daughter would love that!

strong hearth Feb 25, 2024, 11:21 AM

#

https://www.thingiverse.com/thing:4629346 I printed and painted this one

#

Will send I picture of it in a bit

#

boards are ordered 🙂

blissful sluice Feb 25, 2024, 11:22 AM

#

how long did that take to print ?

strong hearth Feb 25, 2024, 11:24 AM

#

it does come in pieces because my build volume is just a little too small. So it took around 2,5 - 3 days of printing.

#

The sanding and painting took over 2 weeks to do lol. That was no fun after 3 days

blissful sluice Feb 25, 2024, 11:28 AM

#

a labour of love... i don't think i would have the patience .... i only really ever print parts for stuff or square boxes to put things in 🙂

strong hearth Feb 25, 2024, 11:33 AM

#

This has been my only big project so far as well. But it's so awesome to see it come to life. So already thinking about next projects xD

blissful sluice Feb 25, 2024, 11:34 AM

#

welcome to the rabbit hole 🙂

strong hearth Feb 25, 2024, 11:35 AM

#

im thinking of printing the arc reactor and making those into voice satellites. That would be awesome

strong hearth Feb 25, 2024, 11:36 AM

#

blissful sluice welcome to the rabbit hole 🙂

happy to be in it lol

blissful sluice Feb 25, 2024, 11:37 AM

#

yeah , it's reall good fun and interesting and great to learn new stuff

strong hearth Feb 25, 2024, 11:38 AM

#

what do you think will be shown in the next HA livestream about Grace?

#Sound effect at wake word detected not playing

- delay: 700ms

speaker: my_speaker