Searching for case insensitive fuzzy query in kanji, katakana and latin

So I have a field - it is a firstName - as such it is mapped as a keyword.
This name can be in kanji, katakana or just latin characters.

So to give an example:
my name is "Markus" - mapped as a keyword

I want to find that when searching for:


So I could do this by making a match query with standard analyzer, fuzziness auto - but when I did that - Kanji searches stopped working. Even when it is the exact Kanji - it didn't find it.

But I can get Kanji to work - by using a fuzzy query - but then it will be case sensitive - so then I can't find MARKUS/markus on the same search.

I am thinking there must be some way of solving this - but I don't know how.

I am using the JAVA Api by the way.

Best regards,

Hi @markuso,

Can you let me know which API (App Search / Enterprise Search) you're running into this issue with, and what version of Enterprise Search you're using?