Skip to main content
RunAnywhere.stt.transcribe takes an AudioInput and returns a Transcription.

Signature

Building an AudioInput

This is the part most people get stuck on. Five constructors:
From a recorded file:
From raw microphone samples:

With options

See STT options for every field.

Checking readiness

Errors

Throws SDKException when no speech model can be loaded or the audio cannot be decoded.