Maia RouterInstall

Summary

Transcribe audio data into text using a selected Maia Router transcription model, providing either a binary audio file from workflow input or an audio URL as input.

Inputs

• Model (required) — Choose between available transcription models ('OpenAI GPT-4o Mini Transcribe' or 'Maia Whisper-1').
• Input Data Mode (required) — Select input mode as either 'Binary File' to use audio from a binary property or 'Audio URL' for a direct audio link.
• Binary Property (required) — Specify the binary property name containing the audio data when 'Binary File' mode is selected (default is 'data').
• Audio URL (required) — Provide the URL of the audio file to transcribe when 'Audio URL' mode is selected.
• Additional Fields — Optionally set parameters such as language, prompt for context, response format (JSON, text, SRT, VTT, verbose JSON), and temperature to control transcription behavior.

Output shape

A single JSON object per input audio item containing the transcription result according to the selected model and response format.

Output structure depends on the chosen 'Response Format' in additional fields; failures return error messages in the JSON output. The operation processes each input item individually.