AQaquavoice
transcription

Transcribe audio synchronously

Uses a 210-second overall deadline. If the wait times out after the job is stored, processing continues and the 504 request_timeout error body carries error.job_id; poll that job to follow its status. The x-request-id response header carries the same value.

post/audio/transcriptions

Request

  • Base URL
    https://api.aquavoice.com/v1
  • Auth
    HTTP bearer

Headers

Idempotency-Keystring

Optional idempotency value. Use the same value of 1–255 printable ASCII characters for retries of one logical request. Terminal results replay for 24 hours, or 10 minutes for a dictation on a zero-data-retention account; different request content returns 409 idempotency_conflict.

X-Client-Request-Idstring

Optional printable ASCII correlation value of up to 512 characters. It does not provide idempotency.

Request body multipart/form-data

filestring binary required

Audio in flac, mp3, mp4, mpeg, mpga, m4a, ogg, opus, wav, or webm format. Maximum 256 MiB and four hours of decoded audio.

model'avalon-v1.5' required

The Avalon model to use.

languagestring

ISO 639-1 language code. Omit it or use auto to detect the language.

promptstring

Context or vocabulary hints for the transcription.

response_format'json' | 'text' | 'srt' | 'vtt' | 'verbose_json' | 'diarized_json'

Response format. Defaults to json, or to diarized_json when diarize is true. A diarized transcription supports only diarized_json, srt, and vtt.

diarizeboolean

Labels who speaks each word. The result has speaker turns with word timings. Speaker IDs (spk_0, spk_1, ...) are in order of first speech. A transcription has a maximum of 8 speakers.

max_speakersinteger

The maximum number of speaker labels. Only with diarize. The speakers with the most detected speech keep their labels; words of other speakers go to the closest kept speaker. It does not change how many speakers the model detects.

temperature0

Compatibility no-op. Nonzero values are unsupported.

streamfalse

Avalon is batch-only, so only false is accepted.

timestamp_granularitiesstring[]

Only segment is supported, and only with verbose_json.

Example request

{
  "language": "auto"
}

Response

Transcription result.

task'transcribe'
textstring required
languagestring
durationnumber
segmentsobject[]

Example response

{
  "text": "The beta is stable enough to expand next week.",
  "language": "en",
  "duration": 17.8,
  "segments": [
    {
      "start": 0,
      "end": 4.2,
      "text": "The beta is stable enough to expand next week."
    }
  ],
  "usage": {
    "seconds": 17.8
  }
}

Changes

No recorded changes to this operation across all 1 revision of this API.