Index analyzer problem with accent!


I want to add asciifolding filter to elasticsearch.yml config file so i added this analyzer :
index :
analysis :
analyzer :
default :
tokenizer : standard
filter : [standard, lowercase, asciifolding]

so when i search for example : "sebast" with prefix or wildcard query, it yield the good doc which contain "sébastien"
however, when i search with "sébast" keyword, with the previous queries dsl, it yield nothing except if i use "query_string", or i need to use wildcard.

So, is there a solution??