Actions2
- Image Actions
- Audio Actions
Audio → Transcribe
AI-generatedSummary
Transcribes each input audio file by sending its URL or base64 data URI to Zihin’s audio transcription API. Requires Zihin API credentials; the request uses the configured base URL and timeout options.
Inputs
- Audio (required) — URL or base64 data URI of the audio file.
- Additional Options → Language — Defaults to Auto Detect. The source sends this value as
languageonly when it is notauto; the other listed choices appear document-oriented. - Options → Base URL Override — API base URL; defaults to
https://llm.zihin.ai. Custom domains require enabling Allow Custom Base URL. Localhost requires Allow Localhost. - Options → Timeout — Request timeout in milliseconds.
0uses the automatic transcription timeout of 395 seconds; positive values are capped at 900,000 ms.
Output shape
One output item per input item, containing the transcription fields returned in the API response’s data object, plus _meta with success and execution_time_ms.
The source only uses the audio value and the Language option from Additional Options; Extract Mode, Sheet Name, and Max Pages are ignored for transcription. API or request errors stop execution unless Continue On Fail is enabled, in which case the affected item contains
errorandsuccess: false.