API Reference
Generate Speech
API Reference
Generate Speech
Synthesize speech asynchronously.
Voice can be specified by providing either voice + model_type, or
voicemodel_uuid.
Pace, duration, and pitch are only supported for voices
that support these controls (a very small subset of voices).
You can check if a given voice supports controls by querying /voice-data or
/voices/<voicemodel-uuid>/detail and looking at the controls boolean in
the reponse.
POST
/
speak
Authorization
Header
Body
curl --request POST \
--url https://api.uberduck.ai/speak \
--header 'Authorization: <authorization>' \
--header 'Content-Type: application/json' \
--data '{
"duration": [
123
],
"model_type": "<string>",
"pace": 123,
"pitch": [
123
],
"speech": "<string>",
"voice": "<string>",
"voicemodel_uuid": "<string>"
}'{
"uuid": "<string>"
}Authorizations
Authorization
string
headerrequiredThe access token received from the authorization server in the OAuth 2.0 flow.
Headers
uberduck-id
string
default: anonymousBody
application/json
duration
number[]
model_type
string
pace
number
default: 1pitch
number[]
speech
string
requiredvoice
string
default: ljvoicemodel_uuid
string
Response
200 - application/json
uuid
string
requiredcurl --request POST \
--url https://api.uberduck.ai/speak \
--header 'Authorization: <authorization>' \
--header 'Content-Type: application/json' \
--data '{
"duration": [
123
],
"model_type": "<string>",
"pace": 123,
"pitch": [
123
],
"speech": "<string>",
"voice": "<string>",
"voicemodel_uuid": "<string>"
}'{
"uuid": "<string>"
}