latest2026.01.16 DevSpeech Service

Create Speech

Creates a new Speech resource. HTTP Mapping: POST /v1/conversation/speech

Query parameters

speechIdstringOptional

speech_id is a client-assigned unique identifier for the Speech resource. This ID is used to reference the speech in subsequent operations.

Request

This endpoint expects an object.
displayNamestringRequired
display name contains the display name for the relevant speech.
contentstringRequired
content contains the actual textual content to be converted into speech. This field is required to generate the speech output.
serviceintegerRequired
service specifies the speech processing service to use for generating the speech. It must be one of the defined values in the SpeechProcessingService enum.
languageTagobjectRequired
language specifies the language of the speech content.
namestringOptional

name is the resource identifier for the Speech. It follows the format “speeches/{speech}” and uniquely identifies a Speech resource.

speeddoubleOptional
speed defines the playback speed of the generated speech. This optional field allows for customization of the speech delivery rate.
outputFormatintegerOptional

output_format determines the file format for the synthesized speech audio, such as MP3 or WAV. This field is optional and depends on the capabilities of the selected service.

typeintegerOptional
type specifies the storage behavior of the speech output. It indicates whether the generated speech should be persisted for future reference or treated as transient.

Response

OK
displayNamestring
display name contains the display name for the relevant speech.
contentstring
content contains the actual textual content to be converted into speech. This field is required to generate the speech output.
serviceinteger
service specifies the speech processing service to use for generating the speech. It must be one of the defined values in the SpeechProcessingService enum.
languageTagobject
language specifies the language of the speech content.
namestring

name is the resource identifier for the Speech. It follows the format “speeches/{speech}” and uniquely identifies a Speech resource.

speeddouble
speed defines the playback speed of the generated speech. This optional field allows for customization of the speech delivery rate.
outputFormatinteger

output_format determines the file format for the synthesized speech audio, such as MP3 or WAV. This field is optional and depends on the capabilities of the selected service.

durationstringRead-onlyformat: "^-?(?:0|[1-9][0-9]{0,11})(?:\.[0-9]{1,9})?s$"
duration specifies the maximum length of the speech output in seconds. This field is required to limit the length of the generated speech.
audioUristringRead-only
Storage URI for the synthesized speech audio. This field is optional and depends on the capabilities of the selected service.
typeinteger
type specifies the storage behavior of the speech output. It indicates whether the generated speech should be persisted for future reference or treated as transient.

Errors

400
Bad Request Error