Speech to text

POST

/v1/audio/transcriptions

Transcribes an audio file into the spoken language using the deployed ASR model.

Request

Send multipart/form-data — do not use application/json for this endpoint.

Field	Type	Required	Notes
`file`	file	Yes	Audio file. Supported formats and the maximum file size depend on the deployed ASR model (commonly `flac`, `mp3`, `mp4`, `mpeg`, `mpga`, `m4a`, `ogg`, `wav`, `webm`).
`model`	string	Yes	An ASR model id from your catalog, e.g. `myt/myt_asr`.
`prompt`	string	No	Optional context or spelling hints to guide the model.
`response_format`	string	No	`json` (default), `text`, and other formats where the model supports them (e.g. `srt`, `verbose_json`, `vtt`).
`temperature`	number	No	Sampling temperature, where supported.

Example request (curl)

curl -sS --request "POST" \
  --url "$MYT_BASE_URL/audio/transcriptions" \
  --header "Authorization: Bearer $MYT_API_KEY" \
  -F "model=myt/myt_asr" \
  -F "file=@/path/to/recording.wav"

Response (`verbose_json` format, where supported)

{
  "task": "transcribe",
  "language": "english",
  "duration": 4.32,
  "text": "The quick brown fox jumps over the lazy dog.",
  "words": [
    { "word": "The",   "start": 0.00, "end": 0.18 },
    { "word": "quick", "start": 0.18, "end": 0.40 }
  ]
}

Errors

Status	When
`400`	File missing, unsupported format, or exceeded size limit
`401`	API key missing or invalid
`429`	Rate limit or budget exceeded

Code examples

curl -sS --request "POST" \
  --url "$MYT_BASE_URL/audio/transcriptions" \
  --header "Authorization: Bearer $MYT_API_KEY" \
  -F "model=myt/myt_asr" \
  -F "file=@/path/to/audio.wav"

Full parameter reference: Audio transcriptions.