Transcription HESS icon

Transcription HESS

Transcrit un fichier audio via le moteur Whisper HESS

Transcription HESS

AI-generated

Summary

Transcribes each input item’s audio file by sending it to the Whisper HESS API. Returns the API response as that item’s JSON data.

Inputs

  • Propriété binaire (required) — Name of the input item’s binary property containing the audio file; defaults to data.
  • Langue — Select French (fr, default) or automatic language detection (auto).
  • Inclure les segments horodatés — Whether to include timestamped segments in the API response; defaults to true.
  • Utiliser le lexique métier — Whether to use the business lexicon to bias decoding; defaults to true.
  • Options — Optional API settings: beam search width from 1 to 10 (default 5 when supplied) and a Whisper model override (blank leaves the API default).

Output shape

A list of items, one per input item, each containing the API response in json.

Requires Transcription HESS API credentials. The audio is sent as multipart form data to the configured base URL’s /transcrire endpoint. With Continue On Fail enabled, an item that errors returns json.error instead of stopping execution.

Links

Discussion