> ## Documentation Index
> Fetch the complete documentation index at: https://docs.vmeg.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# API overview

> What VMEG Open API does and how endpoints are organized

VMEG Open API lets you automate what users do on [vmeg.ai](https://www.vmeg.ai/home) from your own backend: **translate and dub** video or audio, **synthesize speech**, **clone voices**, and **translate plain text**.

<Info>
  **Video/audio translation = translate + dub.** The API transcribes speech, translates the script, generates new AI-dubbed audio (and video when applicable), and returns dubbed media files. It is **not** the same as the text-only translation API.
</Info>

## URLs

| Purpose                                   | URL                                          |
| ----------------------------------------- | -------------------------------------------- |
| Documentation site                        | [https://docs.vmeg.ai](https://docs.vmeg.ai) |
| API base URL (all `/openapi/**` requests) | `https://api.vmeg.ai`                        |

The interactive Playground sends requests to `https://api.vmeg.ai`.

## Authentication

All `/openapi/**` endpoints require your API Key in the **`Authorization` request header**:

```http theme={null}
Authorization: Bearer <api_key>
```

Enter the key in the Playground **Authorization** field to try requests.

## Scope

List, detail, and delete operations for **tasks** and **materials**, and queries for **cloned voices**, return only data tied to the API Key you authenticate with. You do not pass a project identifier in query parameters or request bodies.

## Typical translate-and-dub flow

Most integrators start here when localizing video or audio:

<Steps>
  <Step title="Upload your file">
    Finish the full **Upload files (materials)** flow to get a registered `materialId` for your video or audio. See [Material upload](/guides/assets/material-upload). Do not create a job with a mid-upload `materialId`.
  </Step>

  <Step title="Configure webhook">
    Set your webhook URL on [API Configuration](https://www.vmeg.ai/open-api-setting) so dubbed results can be delivered when the job finishes.
  </Step>

  <Step title="Create translate and dub job">
    `POST /openapi/v1/task/media-translation/create-async` — VMEG transcribes, translates, dubs, and renders output. See [Create translate and dub job](/api-reference/media-translation/create-media-translation-async).
  </Step>

  <Step title="Receive dubbed media">
    Your webhook receives CDN URLs for the dubbed video or audio. Poll **Query job status** if you need a fallback.
  </Step>
</Steps>

## Endpoint groups

| Group                               | What it does                                                                | When to use                                                                  |
| ----------------------------------- | --------------------------------------------------------------------------- | ---------------------------------------------------------------------------- |
| **Video & audio translate and dub** | Full localization pipeline: ASR → translation → AI dubbing → rendered media | You have a video or audio file and want a dubbed version in another language |
| **Text to speech (TTS)**            | Turn text into speech with preset or cloned voices                          | You already have a script and need audio only                                |
| **Voice cloning**                   | Clone a speaker from a short sample                                         | You need a custom `voiceId` for TTS or dubbing                               |
| **Text translation (text only)**    | Translate plain text segments                                               | You have strings only — **not** spoken content in media                      |
| **Task status and history**         | List, query, or delete jobs                                                 | Track progress, recover from missed webhooks                                 |
| **Upload files (materials)**        | Upload and manage video/audio inputs                                        | Required before translate-and-dub jobs                                       |
| **Voices (presets and clones)**     | Browse voices available for TTS                                             | Pick a `voiceId` before synthesizing speech                                  |

Async **`create-async`** endpoints document the **Callbacks** webhook request body (task-specific `data` schema) on the same page.

Concept guides (idempotency, webhooks, errors, pricing) are under the **Guides** tab. Product walkthroughs: [Media translation](/guides/products/media-translation), [Products overview](/guides/products/overview).