/transcribeSynchronously transcribe an audio file.
Transcribe Audio
Uploads an audio file via multipart/form-data and returns the transcription synchronously. The request body is capped at 8 MiB; larger uploads return 413. Empty files return 400. Calls are metered against your account and deducted from credits when transcription completes successfully.
Request body
4filebinaryrequiredAudio file uploaded as
multipart/form-data. Accepted formats: WAV, FLAC, Ogg/Opus. Maximum 8 MiB. Other file types are rejected.languagestringoptionalLanguage hint (e.g.
english,french). Omit to let the model detect the language.group_idstring (uuid)optionalOptional UUID correlating multiple chunked uploads of one recording. Send the same value with every chunk; dashboards aggregate the chunks into a single transcription entry.
group_totalintegeroptionalExpected number of chunks in the group, 1–10000 (requires
group_id). Lets dashboards flag a group whose chunks did not all arrive as partial.
Errors
6The uploaded file is empty (zero bytes).
Upload a non-empty audio file.
Missing or invalid API key.
Add X-API-Key: YOUR_KEY to your request headers.
Your API key does not have the transcribe:use scope.
Issue a new key with the transcribe:use scope.
The uploaded file exceeded the 8 MiB cap.
Compress or trim the audio to fit under 8 MiB before retrying. The presigned-upload flow has the same cap.
Missing file field or unsupported audio format.
Attach a file field with a supported audio format (WAV, FLAC, Ogg/Opus).
The ASR service is temporarily unavailable.
Retry after a few seconds; the service should recover automatically.
