Skip to content

Audio Transcriptions

Transcribes an audio file you upload or one downloaded from a URL. Billed per minute.

Updated on Aug 10, 2026

Transcribes an audio file. Accepts the file in multipart/form-data or a public URL in parameters.audio. Billed per minute of audio. The guide with examples and limits is in Audio transcription.

POSThttps://api.hinow.ai/v1/audio/transcriptionsBearer

Converts speech into text. Synchronous response.

Parâmetros

  • modelstring· bodyobrigatório

    `hinow/hivox`.

  • filefile· body

    The audio, in `multipart/form-data`. Required when there is no URL.

  • parameters.audiostring· body

    Public URL of the audio, in a JSON request. Alternative to `file`.

Respostas

200Transcript and audio duration in seconds.
{
  "duration": 8.5435,
  "text": "Bom dia, Maria. Tudo bem? Tudo, obrigada. E você? Também estou bem. Obrigado. Tchau, João."
}
400`REQUEST_MISSING_FIELD`, `INVALID_MODEL_CATEGORY`, `INVALID_AUDIO_UPLOAD`, or `MISSING_AUDIO`.
502`PROVIDER_FAILED` — unsupported format or a URL that could not be downloaded.

Formats

AcceptedRefused
.mp3, .wav, .ogg.m4a (convert first)

The response has two fields

text and duration. There is no detected language, no timestamps, no speaker separation, and no cost field — billing is calculated from duration.

Was this page helpful?