QOqomplement
voice

Transcribe Audio

Changed on

Transcribe an uploaded audio clip via OpenAI Whisper.

Returns {"text": "<transcribed text>"}. Errors surface as standard HTTP responses so the frontend can show a user-friendly toast.

post/api/voice/transcribe

Request

  • Base URL
    https://api.qomplement.com (the document declares no server; this is the origin it was published from)
  • URL
    https://api.qomplement.com/api/voice/transcribe
  • Auth
    HTTP bearer

Request body multipart/form-data

audiostring binary required

Audio clip from the chat composer mic.

languagestring nullable

ISO-639-1 hint for Whisper. Optional; Whisper auto-detects.

promptstring nullable

Optional context string passed to Whisper so domain vocabulary (product names, connector names like 'Qomplement', 'Kirbi') gets transcribed correctly. Capped at 200 chars server-side.

Response

Successful Response

object required

Changes