OpenAIRealtimeSTTConfig

Defined in: src/providers/stt/openai/OpenAIRealtimeSTT.ts:110

Configuration options for the OpenAIRealtimeSTT provider.

Remarks

Extends STTProviderConfig with OpenAI Realtime-specific settings. You must provide either apiKey (for direct connections) or proxyUrl (for a server-side proxy that injects the API key). If both are provided, proxyUrl takes precedence.

For direct browser connections, mint an ephemeral client secret server-side (POST /v1/realtime/client_secrets) and pass an async factory as apiKey so a fresh secret is fetched on each connection.

Example

// Direct connection with an ephemeral client-secret factory
const config: OpenAIRealtimeSTTConfig = {
  apiKey: async () => {
    const res = await fetch('/api/openai-client-secret');
    const { value } = await res.json();
    return value;
  },
  model: 'gpt-4o-mini-transcribe',
};

// Proxy connection (recommended for production)
const config: OpenAIRealtimeSTTConfig = {
  proxyUrl: 'http://localhost:3001/api/proxy/openai-realtime',
  language: 'en',
};

See

OpenAIRealtimeSTT for the provider class

Extends

• STTProviderConfig

Properties