phonetic token filter plugin

The phonetic token filter takes the following settings:

encoder

Which phonetic encoder to use. Accepts metaphone (default), double_metaphone, soundex, refined_soundex, caverphone1, caverphone2, cologne, nysiis, koelnerphonetik, haasephonetik, beider_morse, daitch_mokotoff.

replace

Whether or not the original token should be replaced by the phonetic token. Accepts true (default) and false. Not supported by beider_morse encoding.

				PUT phonetic_sample
					{
  "settings": {
    "index": {
      "analysis": {
        "analyzer": {
          "my_analyzer": {
            "tokenizer": "standard",
            "filter": [
              "lowercase",
              "my_metaphone"
            ]
          }
        },
        "filter": {
          "my_metaphone": {
            "type": "phonetic",
            "encoder": "metaphone",
            "replace": false
          }
        }
      }
    }
  }
}
				GET phonetic_sample/_analyze
					{
  "analyzer": "my_analyzer",
  "text": "Joe Bloggs"
}
		

1. Returns: J, joe, BLKS, bloggs

It is important to note that "replace": false can lead to unexpected behavior since the original and the phonetically analyzed version are both kept at the same token position. Some queries handle these stacked tokens in special ways. For example, the fuzzy match query does not apply fuzziness to stacked synonym tokens. This can lead to issues that are difficult to diagnose and reason about. For this reason, it is often beneficial to use separate fields for analysis with and without phonetic filtering. That way searches can be run against both fields with differing boosts and trade-offs (e.g. only run a fuzzy match query on the original text field, but not on the phonetic version).

If the double_metaphone encoder is used, then this additional setting is supported:

max_code_len

The maximum length of the emitted metaphone token. Defaults to 4.

If the koelnerphonetik or haasephonetik encoder is used, then this additional setting is supported:

max_variations

Available in 9.4.9, 9.5.6, 9.6.1, and later. The maximum number of spelling variations the encoder generates for a single token. These encoders emit a code for every spelling variation of a token (for example AUN and OWN) and for every combination of its punctuation-separated segments, which grows quickly for long tokens. A token that would exceed the limit is encoded from its first max_variations variations only. Defaults to 256. The memory needed to encode one token grows in proportion to max_variations multiplied by the token length, so raising it on a field whose tokenizer emits long tokens (for example keyword) lets a single large value consume a large amount of heap.

If the beider_morse encoder is used, then these additional settings are supported:

rule_type

Whether matching should be exact or approx (default).

name_type

Whether names are ashkenazi, sephardic, or generic (default).

languageset

An array of languages to check. If not specified, then the language will be guessed. Accepts: any, common, cyrillic, english, french, german, hebrew, hungarian, polish, romanian, russian, spanish.