Transcribe audio · OpenAI / Whisper

Node catalog Artificial intelligence
On this page

Turn speech into text with Whisper or OpenAI GPT transcription models.

Usage and specifics

Select an OpenAI credential and a model suitable for transcription or speech synthesis. Supply the required audio or text and preserve output format and MIME type.

Technical type : transcribe_audio_openai. Runtime : nodulz.

Credentials and prerequisites

Create credentials in your nodulz instance’s Credentials area, then select them on the node. Availability depends on ownership and sharing. See the credentials guide for access and rotation.

Parameters

Names below are technical keys. Defaults are not a ready-to-run configuration: complete fields required by your operation. Paths containing [] describe list items; some fields depend on the selected mode.

Parameter Type Default Values / limits
input string "" -
credentialId string "" -
timeoutSeconds number 120 -; ≥ 1, ≤ 3600
model string "gpt-4o-mini-transcribe" -; min 1
inputFormat enum "mp3" mp3, wav, mp4, mpeg, mpga, m4a, webm, flac, ogg
language string "" -
prompt string "" -

Configuration help

credentialId
AI credential
OpenAI API credential accessible to the workflow owner. The provider bills API requests.
input
Audio source
HTTP(S) URL, base64, data URL or complete HTTP or audio node output. Local files are not accepted.
model
Model
Transcription: choose whisper-1 for Whisper, gpt-4o-mini-transcribe (default), or gpt-4o-transcribe. You can also enter another model ID. Speech: gpt-4o-mini-tts by default.
language
Language
Two-letter language code, such as fr or en. Leave empty for automatic detection.

Initial configuration

Initial catalog configuration example, to adapt before running. Keep secrets in credentials.

json
{
  "input": "",
  "credentialId": "",
  "timeoutSeconds": 120,
  "model": "gpt-4o-mini-transcribe",
  "inputFormat": "mp3",
  "language": "",
  "prompt": ""
}

Verification

Test with representative data. Check node output in the inspector and logs, then use editor mapping for downstream fields. A trigger test with a simulated payload does not validate external service delivery.

See also : Expressions and data, Executions, Node catalog.