Class OpenAISTTProvider

Default STT implementation using OpenAI's gpt-4o-transcribe model.

Chunks audio exceeding the per-request limit; chunks are transcribed independently and concatenated with single spaces.

Implements

STTProvider

Index

Constructors

Properties

model transcribeLimitSeconds

Methods

Constructors

constructor

new OpenAISTTProvider(options?: OpenAISTTProviderOptions): OpenAISTTProvider
Parameters
- options: OpenAISTTProviderOptions = {}
Returns OpenAISTTProvider
- Defined in work/scenario/scenario/javascript/src/voice/stt/openai-stt.ts:51

Properties

`Readonly`model

model: string

`Readonly`transcribeLimitSeconds

transcribeLimitSeconds: number

Methods

transcribe

transcribe(audio: AudioChunk): Promise<string>
Parameters
- audio: AudioChunk
Returns Promise<string>
Implementation of STTProvider.transcribe
- Defined in work/scenario/scenario/javascript/src/voice/stt/openai-stt.ts:58