I have an index with Name field . I want to use soundex analyzer and synonym analyzer on that field.

I want to achieve both in a single index .Is it even possible ? Please help me experts out there

Index 1

{
"settings": {
"index": {
"number_of_shards": "1",
"provided_name": "phonetic_sample",
"creation_date": "1603097131476",
"analysis": {
"filter": {
"my_soundex": {
"replace": "false",
"type": "phonetic",
"encoder": "soundex"
}
},
"analyzer": {
"my_analyzer": {
"filter": [
"lowercase",
"my_soundex"
],
"tokenizer": "standard"
}
}
}

I query for Catherine and match Catherine,Katherine and Kathryn

Index 2

{
"settings": {
"index": {
"number_of_shards": "1",
"provided_name": "phonetic_synonym",
"creation_date": "1603121439096",
"analysis": {
"filter": {
"synonym": {
"format": "wordnet",
"type": "synonym",
"synonyms": [
"s(100000001,1,'Bill',v,1,0).",
"s(100000001,2,'William',v,1,0).",
"s(100000001,3,'Wilhelm',v,1,0)."
]
}
},
"analyzer": {
"synonym": {
"filter": [
"synonym"
],
"tokenizer": "whitespace"
}
}
}

I query for Bill and match Bill, William and Wilhelm

You can use multi-field with multiple analyzers. You can declare sub-fields for the name field, each with a different analyzer.

Below is the modified index mapping.

Index Mapping:

{
  "settings": {
    "index": {
      "analysis": {
        "filter": {
          "my_soundex": {
            "type": "phonetic",
            "encoder": "metaphone",
            "replace": false
          },
          "synonym": {
            "format": "wordnet",
            "type": "synonym",
            "synonyms": [
              "s(100000001,1,'Bill',v,1,0).",
              "s(100000001,2,'William',v,1,0).",
              "s(100000001,3,'Wilhelm',v,1,0)."
            ]
          }
        },
        "analyzer": {
          "synonym": {
            "filter": [
              "synonym"
            ],
            "tokenizer": "whitespace"
          },
          "my_analyzer": {
            "filter": [
              "lowercase",
              "my_soundex"
            ],
            "tokenizer": "standard"
          }
        }
      }
    }
  },
  "mappings": {
    "properties": {
      "name": {
        "type": "text",
        "analzyer": "synonym",
        "search_analyzer": "synonym",
        "fields": {
          "content": {
            "type": "text",
            "analyzer": "my_analyzer",
            "search_analyzer": "my_analyzer"
          }
        }
      }
    }
  }
}

Then you can refer to name and name.content in your queries. Your search query will be like this:

{
  "query": {
    "multi_match": {
      "query": "Bill",
      "fields": [ 
        "name",
        "name.content"
      ],
      "type": "most_fields" 
    }
  }
}