I haven’t tried adding timers, but everything else on the Box and Sensor Dock work well and is up to date with the latest version. Here’s my config:
1 Like
Anyone have a solution to be able to control the volume on these things? The one custom ESPhome component I found doesn’t work anymore, and the default volume is WAY too quiet!
See the post directly above. 
My github has working volume control. BigBobba’s does too.
1 Like
After flashing one of my Box3s with BigBobbas and trying to use their yaml and AlmostInteractive’s yaml, I’m eventually able to get it up and running. I’ve got the extra sensors and even got the media player/volume working—all great things.
However, the actual voice assistant bits seem to be 99% non-functional - They pick up the wake word but (again 99% of the time) just go from listening to the confused look by a flash of an orange icon and back to ‘listening’ – Since there is not feedback I have no idea what’s broken and the logs show absolutely nothing out of the ordinary.
Note - my stock Box3 (and this one with stock firmware) work just fine - it just means I lose all extended functionality, including volume, which seems to be everyone’s largest complaint (mine included).
I realize I’m giving useful detail about F.all here, and I know I need to go digging, but I was hoping that I might not be an outlier and that someone here might have an answer before I actually get time to go digging.
try changing the ‘finished speaking detection’ setting in the HA integration for the device. Also if using my config then try the config in the dev branch relevant to your device, if you have any probs raise and issue in the repo or join us on discord … link can be found in the github ‘discussions’
Thanks for your support by trying the project
Anybody got the battery percentage to work correctly? When it reaches 60% or so there’s a send drop.
Is there a BigBobba’s version for the Lite?
Hi ,
the project should work on the lite using any of the below.
if you use the no_sensor version 
Main Branch here
Dev Branch here
WIP Branch here
Latest with Alarmo support here
all info and details can be found here
or if you have any questions you can join us on Discord
Cheers
5 Likes
@robgough1970 thank you for your project and quick help on Github. Is it possible to change a pictures of the Assist face to custom, like a main S3ox tutorial mention? My custom voice is Capitan Morgan 
and wake word is Hey, Morgan, so I want to put pirate bird picture instead of the house. Thank you
Thanks so much for your answer and your work.
Definitely needs some flexibility. People in my household are used to saying: “Alexa, 10 minutes” to start a timer.
That would be rather difficult don’t you think so. There is absolutely no context in that request. Even a human will have to guess is it is a timer, 10 minutes walking from geofence, agenda point…?!?
But if you are sure that this always would be a timer then just at it to an automation or include it into your custom sentences.
Try the 3(B) version - same did happen for me…
Hi Tobias, what was your ultimate solution for broadcasting the S3 box audio? Did you get it to work BT or did you create a 3.5mm audio out jack component?
I’m trying to determine a solution for my own S3 box and want to output the S3 sound to something external so I can amplify it and broadcast it to external speakers.
Having difficulties on my end routing the S3 box audio to anything external.
Would you mind sharing your code on where and what you added? Thx!
Many thanks!! Only just back home, will give that a shot today/tomorrow.