train on names
We have 2 users with the same first name. The ai receptionist gets lost when trying to decipher between the two. Please make a way to train the ai on specific names

-
Hi William and Pete,
Thanks for the suggestion. I am the PM working on AIR, and this is very good feedback.
We have a few suggestions that can help:
* If there are multiple users with the same first name, AIR will usually either mention their full names (if available) or ask the caller to clarify what their full name is.
* In addition to this, if more info about the user is made available in the User directory, such as their department or job title, AIR can use that as well to correctly identify the person.
* We use fuzzy search that is more lenient at understanding names and tries to match with the correct user, but there are always opportunities for it to learn and do better.
* The point which Pete mentioned about alternate name - that is also something we can support if that info is entered into the User directory.Thanks,
Deepak -
Pete commented
I found a workaround for my Suzy/Susie problem. I don't know if it would help others, but I used Transfer on Context and put "Susie" in the context and directed it to Suzy's extension. IT WORKED!
-
Pete commented
This is a REALLY important to the functionality of the product. I completely agree with William's post is important and should be implemented urgently. I'd like to add another use case. We're a small company and have unique user names so people call and ask by first name. AIR appears to transcribe the voice making the request, and then search for user names that match. However, our user's name is "Suzy" and AIR translates that to "Susie" and is unable to find it. This is also an occasional issue with "Rachael" being spelled differently, apparently depending on the caller's pronunciation. Even "Peter" and "Pete" aren't handled well. While I'm not an AIR developer, my thought would be to add a nickname or alternate name table that would be searched first to reconcile the user name. This issue is critical to our continued use of AIR.