Translations

Translate non-English audio into English text with the OpenAI Audio Translations API.

Upload a non-English audio file and translation parameters to receive English text. Use the Transcriptions API when you need the transcript in the original language.

POST/v1/audio/translations

Upload non-English audio files and translate the audio content into English text.

Authorization

bearerAuth

AuthorizationBearer <token>

Use Bearer your-api-key in the Authorization Header.

In: header

Request Body

multipart/form-data

Use Multipart to submit audio files and translation parameters. The client should automatically generate the boundary of the Content-Type from the request library.

TypeScript Definitions

Use the request body type in TypeScript.

file*file

The audio file to be translated. Supports FLAC, MP3, MP4, MPEG, MPGA, M4A, OGG, WAV, and WebM formats.

Formatbinary

model*string

The audio translation model ID to be used, such as whisper-1. The actual available models are based on the model list returned by the current API Key.

Length1 <= length

prompt?string

English text to guide the model's style.

response_format?string

Output format of the translation results.

Default"json"

Value in

"json"
"text"
"srt"
"verbose_json"
"vtt"

temperature?number

Sampling temperature. Higher values make the output more random.

Range0 <= value <= 1

Default0

Response Body

`application/json`

curl -X POST "https://example.com/v1/audio/translations" \  -F file="chinese.mp3" \  -F model="MODEL_ID"

{
  "text": "Hello, world!"
}

Authorization

Request Body

Response Body

200application/json

400application/json

401application/json

403application/json

429application/json

500application/json

`application/json`

`application/json`

`application/json`

`application/json`

`application/json`

`application/json`