API
Audio transcription
Transcribe audio files with OpenAI-compatible speech-to-text requests.
Overview
The transcription endpoint converts audio into text. Use multipart form data and provide a model that supports transcription.
Request
`POST https://uouo.cloud/v1/audio/transcriptions`
Form fields
| Field | Required | Description |
|---|---|---|
| file | Yes | Audio file. |
| model | Yes | Transcription model ID. |
| language | No | Optional ISO language hint. |
| prompt | No | Optional context prompt. |
| response_format | No | `json`, `text`, `srt`, `verbose_json`, or provider-supported format. |
Example
curl https://uouo.cloud/v1/audio/transcriptions \
-H "Authorization: Bearer $UOUODUO_API_KEY" \
-F file="@meeting.mp3" \
-F model="whisper-1" \
-F response_format="json"Notes
- Audio size and duration limits depend on the upstream model.
- Avoid sending sensitive recordings unless your product policy allows it.
- For long audio, split files into chunks and stitch results with timestamps.