SpeechModelSpec
Defined in: core/media-types.ts:245
Specification for a speech generation model.
Properties
Section titled “Properties”defaultFormat?
Section titled “defaultFormat?”
optionaldefaultFormat:AudioFormat
Defined in: core/media-types.ts:271
Default format if not specified
defaultVoice?
Section titled “defaultVoice?”
optionaldefaultVoice:string
Defined in: core/media-types.ts:268
Default voice if not specified
displayName
Section titled “displayName”displayName:
string
Defined in: core/media-types.ts:253
Human-readable display name
features?
Section titled “features?”
optionalfeatures:object
Defined in: core/media-types.ts:274
Additional feature flags
languages?
Section titled “languages?”
optionallanguages:number
Number of supported languages
multiSpeaker?
Section titled “multiSpeaker?”
optionalmultiSpeaker:boolean
Supports multi-speaker output
voiceInstructions?
Section titled “voiceInstructions?”
optionalvoiceInstructions:boolean
Supports voice instructions/steering
formats
Section titled “formats”formats:
AudioFormat[]
Defined in: core/media-types.ts:262
Supported audio formats
maxInputLength
Section titled “maxInputLength”maxInputLength:
number
Defined in: core/media-types.ts:265
Maximum input text length (characters)
modelId
Section titled “modelId”modelId:
string
Defined in: core/media-types.ts:250
Model identifier
pricing
Section titled “pricing”pricing:
SpeechModelPricing
Defined in: core/media-types.ts:256
Pricing information
provider
Section titled “provider”provider:
string
Defined in: core/media-types.ts:247
Provider identifier (e.g., “openai”, “gemini”)
voices
Section titled “voices”voices:
string[]
Defined in: core/media-types.ts:259
Available voice options