Voice Interviews

Turn one recorded spoken answer into text for the user to check before it is accepted.

Transcribes one recorded answer and returns the text. Send the recording base64-encoded in answer_audio_base64 with its container in answer_audio_content_type. Show the returned text to the user to confirm or correct before you accept it with advance_voice_interview — that review step is the point of this being a separate call. The recording is transcribed and discarded; it is never stored. Record compactly: a short, compressed clip transcribes several times faster than a large uncompressed one of the same speech.

post/v1/transcribe_voice_interview_answer

Request

  • Base URL: https://partner.mindupload.app
  • URL: https://partner.mindupload.app/v1/transcribe_voice_interview_answer
  • Auth: API key in header X-Partner-Key

Request body

preferred_languagestring nullable

BCP-47-ish language code (e.g. 'en', 'zh-cn') for localized messages.

usernamestring nullable

The end-user's username.

passwordstring nullable

The end-user's password on first sign-in. After login/register, send the returned session token (JWT) here instead — the field accepts either, and reusing the token keeps raw passwords out of your storage.

interview_session_idstring nullable

Identifies one interview. Returned by start_voice_interview and required by every later call.

answer_audio_base64string nullable

The user's recorded answer, base64-encoded. Discarded after transcription; never stored.

answer_audio_content_typestring nullable

Container of the recording being sent.

Response

Result envelope — check the success field.

successboolean

Whether the call succeeded.

error_messagestring nullable

Human-readable reason when success is false.

error_codestring
recognized_textstring nullable

What the user was heard to say. Show it to them to confirm or correct.

Changes

No recorded changes to this operation across all 2 revisions of this API.