Universal Transcription

Generate transcripts from audio files using multiple AI providers

Universal Transcription

AI-generated

Summary

Transcribes audio with a selected Google Generative AI or OpenAI model, using audio from an n8n binary property or a URL. Optionally sets the audio language and requests timestamped segments.

Inputs

  • Provider (required) — Choose Google Generative AI or OpenAI.
  • Audio Source (required) — Choose Binary Data or URL as the audio source.
  • Audio Property (required) — For Binary Data, the name of the input binary property containing the audio; defaults to data.
  • Audio URL (required) — For URL, the address of the audio file to transcribe.
  • Model (required) — Select a transcription model; available models depend on the provider.
  • Options — Optional settings: Language accepts an ISO-639-1 code (blank leaves language detection to the model), and Include Segments requests timestamped segments.

Output shape

A transcription result for each input item.

The provided source pack shows the transcription handler's signature but omits its implementation, so the exact output fields and segment format are not available.

Links

Discussion