I’m confused by the different elements of HA’s voice assistant sentences.
-
What’s the difference between a
conversation
and anintent_script
? Per HA’s custom sentence example, aconversation
has anintents
sub-element, and anintent_script
doesn’t. Does aconversation
’sintent
merely declare the element that will respond to the sentence, while anintent_script
is purely the response (i.e., does anintents
point to anintent_script
)? -
HA then explains that while the example above defined the
conversation
andintent_script
inconfiguration.yaml
, you can also defineintents
inconfig/custom_sentences/
. Should you use both of these methods simultaneously or will it cause conflict or degrade performance? I wouldn’t think you should define the same sentence in both places, but the data structure for their 2 examples are different - is 1 better than the other?
In configuration.yaml
:
conversation: intents: YearOfVoice: - "how is the year of voice going"
In config/custom_sentences/en
:
intents: SetVolume: data: - sentences: - "(set|change) {media_player} volume to {volume} [percent]" - "(set|change) [the] volume for {media_player} to {volume} [percent]"
- Then they say
responses
for existing intents can be customized as well inconfig/custom_sentences/
. What’s the difference between aresponse
and anintent_script
? It seems likeintent_script
can only be defined inconfiguration.yaml
andresponses
can only be defined in config/custom_sentences/` - is that right?
Thanks for any clarification you can share.
RandomLegend@lemmy.dbzer0.com 11 months ago
I’m very confused and interested in an explanation as well
I just setup whisper on my external GPU server to run the medium model with 0.5s of processing time but the built-in intends are somewhat lacking.
mike_wooskey@lemmy.d.thewooskeys.com 11 months ago
What’s involveditn running whisper on a computer other than the home assistant computer? I’m guessing its relatively easyyto install, hopefully in docker. How do you tell HA to use that whisper?
Also, its a bit surprising that moving the voice recognition to a GPU on a more powerful (presumably) computer doesn’t improve HA performance.
RandomLegend@lemmy.dbzer0.com 11 months ago
First of all: It increases performance tremendously. For comparison
So running it on a cheap 100€ used GPU can get you results where Alexa, Siri and Google have to respect you in terms of accuracy and speed. This is a gamechanger for me. I already installed 3 M5Stack ATOM ECHOs in my Home and more will soon come in. It’s incredibly accurate and quick.
Now, to get it running it’s actually pretty easy. First go to this link and download all the files. You have to build a custom docker image with those files. I have no idea how to do that with barebones docker as i am using portainer for everything. In Portainer you have to do:
Next you go
That will spin up a docker-compose with the local custom image you just built, running faster-whisper that is compatible with the wyoming protocol in home assistant and that can run on an NVidia GPU with cuda acceleration.
As you can see in the docker-compose it will expose port 10300. Next:
It will automagically know that it’s whisper and will be fully integrated into your system. You can now add it into your voice assistant.
If you look at the logs of your new docker container you can see every voice command that is sent to your new whisper.