SpeechAudioGenerationInterfaceGenerateSpeechMessage
HTTP API schema
← HTTP API referenceView source specification ↗
Turn text into speech audio using a model and voice from the model catalog.
SpeechAudioGenerationInterfaceGenerateSpeechMessage fields
- model
stringRequired Model identifier from the model catalog (query-models). An unknown model is rejected.
- text
stringRequired The text content to synthesize into speech audio
- voice
string Voice from the model catalog entry. Omit for the vendor default voice.
- styleInstructions
string Delivery direction separate from the spoken text, for models whose catalog entry supports it
- responseFormatBinaryResponseFormat
responseFormat: Reference BinaryResponseFormat