Class OpenAIAudioTextToSpeechCapabilityImpl

Description

Provider capability implementation for OpenAIAudioTextToSpeechCapabilityImpl.

Implements

Index

Constructors

Methods

resolveAudioOutputMimeType textToSpeech textToSpeechStream

Constructors

constructor

new OpenAIAudioTextToSpeechCapabilityImpl(
provider: BaseProvider,
client: OpenAI,
): OpenAIAudioTextToSpeechCapabilityImpl
Creates a new OpenAI TTS capability delegate.
Parameters
- provider: BaseProvider
  Parent provider for lifecycle/config access
- client: OpenAI
  Initialized OpenAI SDK client
Returns OpenAIAudioTextToSpeechCapabilityImpl
- Defined in src/providers/openai/capabilities/OpenAIAudioTextToSpeechCapabilityImpl.ts:57

Methods

resolveAudioOutputMimeType

resolveAudioOutputMimeType(format?: string, header?: string | null): string
Parameters
- Optionalformat: string
- Optionalheader: string | null
Returns string
- Defined in src/providers/openai/capabilities/OpenAIAudioTextToSpeechCapabilityImpl.ts:316

textToSpeech

textToSpeech(
    request: AIRequest<ClientTextToSpeechRequest>,
    _ctx?: MultiModalExecutionContext,
    signal?: AbortSignal,
): Promise<AIResponse<NormalizedAudio[]>>
Synthesizes speech in a single non-streaming request.
Parameters
- request: AIRequest<ClientTextToSpeechRequest>
  Unified AI request containing TTS input/options/context
- Optional_ctx: MultiModalExecutionContext
  Optional multimodal execution context (unused by this capability)
- Optionalsignal: AbortSignal
  Optional abort signal
Returns Promise<AIResponse<NormalizedAudio[]>>
Provider-normalized TTS audio artifact response

Throws
If input text is empty or request is aborted before execution
Implementation of TextToSpeechCapability.textToSpeech
- Defined in src/providers/openai/capabilities/OpenAIAudioTextToSpeechCapabilityImpl.ts:71

textToSpeechStream

textToSpeechStream(
    request: AIRequest<ClientTextToSpeechRequest>,
    _ctx?: MultiModalExecutionContext,
    signal?: AbortSignal,
): AsyncGenerator<AIResponseChunk<NormalizedAudio[]>>
Streams synthesized speech as incremental audio chunks plus a final full artifact.
Parameters
- request: AIRequest<ClientTextToSpeechRequest>
  Unified AI request containing TTS input/options/context
- Optional_ctx: MultiModalExecutionContext
  Optional multimodal execution context (unused by this capability)
- Optionalsignal: AbortSignal
  Optional abort signal
Returns AsyncGenerator<AIResponseChunk<NormalizedAudio[]>>
Async generator of audio chunk deltas and a final terminal chunk

Throws
If input text is empty or unsupported stream format is requested
Implementation of TextToSpeechStreamCapability.textToSpeechStream
- Defined in src/providers/openai/capabilities/OpenAIAudioTextToSpeechCapabilityImpl.ts:144