-
-
Notifications
You must be signed in to change notification settings - Fork 27
Update respeaker-satellite-base.yaml #68
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
base: main
Are you sure you want to change the base?
Conversation
added sensor for phase id's added test sensors for request & response text added a virtual touch user button for remote single click
|
Regarding sensors - I honestly think it's pretty niche idea. Truncating text because of max state length can be ok for your case, but not ok for someone else... |
|
here you can see it running: https://www.facebook.com/share/v/16zmaZoo3T/ and yes the text part could be done both ways, i actually have both ways in my yaml so it's optional which way to use, but sure if you could add event for responses that could be just as good, you only have for tts and ur and trust me, a LOT of people like the screens for their VA's, and clocks 😅 |
|
I understand your use case. It just doesn't make sense for basic integration config, because it will force users to the way it's done. Reading sensors from another ESPHome device isn't default and common thing, so I will stay on my point that additions like this belong to forks, not to main, basic integration. I can leave the PR here open, if you want - so others could use your code. :) |
|
problem is having read the events from an automation and place it in a text helper will also truncate to 255.. anyway i understand you don't see the same potential in "outsourcing" the heavy work to another esp, i see it differently, the more i can expose, the more users can customize, and gives no maintenance since it's up the user to use or not (or how) leaving the doors open..🤷🏼♂️ about the events, maybe you should add for tts too, thats what most would want to put on the dash, not so much the stt? |
|
Yeah I guess TTS URI (for playback) isn't enough, having text is needed sometimes. Regarding truncation - well, you could put it to other ESP with service call, and there you can use whatever symbols count you like. But the core problem isn't truncation itself - but the fact it's done at the source, so basically user cannot decide for themselves what they want to do with full text - because there's no full text anymore. |
|
thats why i use both ways (events and text sensors) 😁 choices, but i will try with service call and events 👍 |
|
i like to make things simple so that everyone can play along where most is setup already, then leave entities for those that want to change the default functionality, like swipes for touch, automatically do certain things, but there's a switch to turn that off and do it internally in HA vith automations. some people put up some cool projects but with a million files and automations and external tools and it just to overwhelming that many people drop even trying it, sadly. make it simple, more will use, make it configurable, even more.. about configurable, surprised there is no switch to turn off the led, omg that light blinds me 😅 |
|
I didn't go deep in search of solution to sounds. Again - this is pretty basic implementation, and I don't want to give people more than basic things. For building up on this foundation we have community - you involved :) LED brightness can be set to minimum. If you want to get rid of it completely - it's pretty easy too, I just never had such a request... :) Also I keep YAML as close to PE as possible, because it's easier to merge new changes this way. So - you don't need to convince me that your case is valid and good. You're right. But it's basically narrow path and just one of many, many possibilities that ESPHome giv3s. So I drew the line to avoid making monstrous creature from this YAML. Simple and extendable. |
|
brightness! thank you for saving my eyes the torture!, -the reason you never heard that request before is that your device don't have a display, but it do now 😁 |
|
i got it working using the events instead of text sensors, but seems like we are still being truncated to 500 did some troubleshooting and after adding this to the yaml just before the event is sent: i get this: [06:41:28][D][voice_assistant:764]: Assist Pipeline ended seems like the text is already truncated when we get it from the pipeline 🤷🏼♂️ |
|
Well that was inevitable... |
|
i would call it surprisingly unexpected :( |
Any network protocol would restrict max data length. |
|
i don't think it's a network restriction but tested some more and can send longer text from developer tools to the recieving device, so thats not where it's truncated along the way, so i need to find the text before the espdevice gets it, i guess..🤷🏼♂️ then again, i didn't think much about the full lengt and that some people might want that for something? (go to debug menu and get it?) but since you mentioned it i kinda got obsessed with doing it 😅 500 characters is just fine (and safe) for small displays and wall dashboards 🤷🏼♂️ sidenote to the "and safe", around 788 characters, above that the display crashes, i believe LVGL and memory, without showing the text (still recieving it) i can send more.. so yeah 500 is safer 👀 |
|
Yeah I guess hardware will limit somewhere there, but I'm surprised that 500 symbols limit is introduced somewhere early. |
|
removed text sensors and only using events now, added node name to events as it's much easier to read than device id (no need to search for id) |
added sensor for phase id's
added text sensors for request & response text
added a virtual touch user button for remote single click
this makes it possible to use another esp to show images for each phase id, show text from respeaker and the virtual button makes it possible to use the touch of remote screen to single click respeaker. (could be expanded to do more than single click)