Implementation:Openai Openai node Translations Create

Knowledge Sources	openai-node OpenAI Translation API
Domains	Audio, Translation
Last Updated	2026-02-15 00:00 GMT

Overview

Concrete tool for translating audio to English text provided by the openai-node SDK.

Description

The Translations.create() method has 4 overloads for different response formats. It uploads an audio file via multipart form to /audio/translations and returns the English translation.

Usage

Use this method to translate non-English audio files to English text.

Code Reference

Source Location

• Repository: openai-node
• File: src/resources/audio/translations.ts
• Lines: L11-42 (class with overloads), L72-109 (TranslationCreateParams)

Signature

class Translations extends APIResource {
  create(
    body: TranslationCreateParams & { response_format?: 'json' },
    options?: RequestOptions,
  ): APIPromise<Translation>;

  create(
    body: TranslationCreateParams & { response_format: 'verbose_json' },
    options?: RequestOptions,
  ): APIPromise<TranslationVerbose>;

  create(
    body: TranslationCreateParams & { response_format: 'text' | 'srt' | 'vtt' },
    options?: RequestOptions,
  ): APIPromise<string>;
}

interface TranslationCreateParams {
  file: Uploadable;
  model: string | AudioModel;
  prompt?: string;
  response_format?: 'json' | 'text' | 'srt' | 'verbose_json' | 'vtt';
  temperature?: number;
}

Import

import OpenAI from 'openai';
// Access via: client.audio.translations.create(...)

I/O Contract

Inputs

Name	Type	Required	Description
file	Uploadable	Yes	Non-English audio file to translate
model	AudioModel	Yes	Model ('whisper-1')
prompt	string	No	English context hint
response_format	string	No (default 'json')	Output format
temperature	number	No	Sampling temperature

Outputs

Name	Type	Description
(json)	Translation	{ text: string } — English translation
(verbose_json)	TranslationVerbose	{ text, language, duration, segments }
(text/srt/vtt)	string	Plain English text or subtitle format

Usage Examples

import OpenAI from 'openai';
import fs from 'fs';

const client = new OpenAI();

// Translate French audio to English
const translation = await client.audio.translations.create({
  file: fs.createReadStream('french_audio.mp3'),
  model: 'whisper-1',
});

console.log('English translation:', translation.text);

// With verbose output
const verbose = await client.audio.translations.create({
  file: fs.createReadStream('german_audio.mp3'),
  model: 'whisper-1',
  response_format: 'verbose_json',
});

console.log('Duration:', verbose.duration);
console.log('Detected language:', verbose.language);
console.log('Translation:', verbose.text);

Related Pages

Implements Principle

• Principle:Openai_Openai_node_Audio_Translation

Requires Environment

• Environment:Openai_Openai_node_Node_20_Runtime