InworldSTTModule | Inherits from: InworldFrameworkModule
Module for Speech-to-Text (STT) functionality within the Inworld framework. Converts audio input into text transcriptions using AI-powered speech recognition. Supports both synchronous and asynchronous speech recognition operations.
Methods
Reference
RecognizeSpeech
Performs synchronous speech recognition on the provided audio chunk. Converts audio data to text using the configured speech recognition model.Parameters
| Parameter | Type | Description |
|---|---|---|
| audioChunk | AudioChunk | The audio data to transcribe into text. |
Returns
Type:string
Description: The transcribed text result, or empty string if recognition failed.
RecognizeSpeechAsync
Performs asynchronous speech recognition on the provided audio chunk. Converts audio data to text using the configured speech recognition model. Provides progress notifications through task events during processing.Parameters
| Parameter | Type | Description |
|---|---|---|
| audioChunk | AudioChunk | The audio data to transcribe into text. |
Returns
Type:Awaitable<string>
Description: A task that completes with the transcribed text result, or empty string if recognition failed.
Configuration Management
Remote Configuration
When using remote STT services, the module creates anSTTRemoteConfig with:
- API key from framework utilities (if available)
Local Configuration
When using local STT models, the module creates anSTTLocalConfig with:
- Model path (resolved from StreamingAssets folder)
- Device configuration from framework utilities