Method: projects.locations.phraseSets.create

Create a set of phrase hints. Each item in the set can be a single word or a multi-word phrase. The items in the PhraseSet are favored by the recognition model when you send a call that includes the PhraseSet.

HTTP request


The URL uses gRPC Transcoding syntax.

Path parameters



Required. The parent resource where this phrase set will be created. Format:


Speech-to-Text supports three locations: global, us (US North America), and eu (Europe). If you are calling the endpoint, use the global location. To specify a region, use a regional endpoint with matching us or eu location value.

Authorization requires the following IAM permission on the specified resource parent:

  • speech.phraseSets.create

Request body

The request body contains data with the following structure:

JSON representation
  "phraseSetId": string,
  "phraseSet": {
    object (PhraseSet)


Required. The ID to use for the phrase set, which will become the final component of the phrase set's resource name.

This value should restrict to letters, numbers, and hyphens, with the first character a letter, the last a letter or a number, and be 4-63 characters.


object (PhraseSet)

Required. The phrase set to create.

Response body

If successful, the response body contains a newly created instance of PhraseSet.

Authorization Scopes

Requires the following OAuth scope:


For more information, see the Authentication Overview.