ModelSpeechConfiguration
HTTP API schema
← HTTP API referenceView source specification ↗
Text to speech configuration.
ModelSpeechConfiguration fields
- voices
array<string> Voices the model offers.
- Item
string
- Item
- defaultVoice
string The voice the vendor uses when `voice` is omitted, when the vendor documents one.
- customVoicesAccepted
boolean True when `voice` also accepts a voice name that is not in `voices`.
- soundTags
array<string> Emotion or sound tags the model documents, written exactly as they appear inline in the text, for example [whispering] or <laugh>. The list is documented examples, not a filter; the model may understand more.
- Item
string
- Item
- supportsStyleInstructions
boolean True when `styleInstructions` steers the delivery.