extended_openai_conversation
extended_openai_conversation copied to clipboard
convert the username
Hi! I am glad that you can attach username in the chat. Unfortunately this is not possible for us, because our name contains Hungarian (European) characters (etc.: István Pintér) If I turn it on, I get the following error:
Sorry, I had a problem talking to OpenAI: 'Pintér István' does not match '^[a-zA-Z0-9_-]{1,64}$' - 'messages.1.name'
Is character encoding correct?
Thanks for your work! Happy New Year!
Thanks for reporting an issue!
Currently, it uses friendly_name
of person
entity to attach.
I might have to change it to something else such as user_id
.
Any suggestions are welcome.
@Teagan42 What's your thought? any ideas?
How about using entity_id
? since any logged user have their own person entity and we commonly use friendly_name
and entity_id
in exposed_entities
, I think using entity_id
can reduce the steps OpenAI should figure out who the speaker is. All entity_id
is written in common Alphabet.
And If we use entity_id
, we might connect speaker recognition function to room presence functionality in near future.
@saya6k
That's a good idea.
I worried that using user_id
might make it hard to read for humans, but using entity_id
will solve this problem!
Thanks for your suggestion!
Sorry, didn't see this thread. I think the user's id and a function to retrieve the user's detail would be a better solution than the one I had implemented.
No problem!
If we use entity_id
(such as person.xxxxxx
), retrieving users' detail will be possible through get_attributes function.
Sorry for being late, and it's released in 1.0.4
.
Please reopen the issue for further issues.