CORTEX
Sign in

Utilities

One-shot helpers that don't fit a domain — currently audio transcription.

Common responses

These responses apply across endpoints in this section unless an endpoint documents an additional response.

StatusMeaningBody
200Request succeeded.Endpoint response object.
201Resource created.Created resource object.
400Invalid request.Error envelope.
401Missing or invalid bearer token.Error envelope.
404Resource not found or not visible to the current organization.Error envelope.

Transcribe

Transcribe audio

Transcribe an audio clip to text using the platform's speech model (OpenAI Whisper). Send a `multipart/form-data` request with a single field named `audio` containing the audio file. Accepted formats: webm, ogg, wav, mp3, m4a — anything Whisper accepts. Maximum file size is 25 MB. Returns `{ text }` with the transcribed content. Returns 503 if the server is not configured with an OpenAI API key.

POST/v1/transcribe

Request body

NameTypeRequiredDescription
audiofileRequiredAudio file to transcribe. `multipart/form-data` field name must be `audio`. Accepted MIME types: audio/webm, audio/ogg, audio/wav, audio/mpeg, audio/mp4. Maximum 25 MB.
Example requestbash
curl https://api.cortex.cognit-dx.com/v1/transcribe \
  -X POST \
  -H "Authorization: Bearer $CORTEX_TOKEN" \
  -F audio=@/path/to/audio.webm

Response codes

StatusMeaningBody
401Missing or invalid credential.`{ error, code, request_id }`
422Request body failed validation.`{ error, code, request_id }`
200 OKjson
{
  "text": "Schedule a follow-up call with the Acme team for next Tuesday at 10 AM."
}