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

# Training from an Image

> Upload a headshot and select a voice_name to create a replica without recording a training video.

Use this path when you call [Create Replica](/api-reference/phoenix-replica-model/create-replica) with **`train_image_url`** and **`voice_name`**. The image file must be reachable at a **publicly accessible URL** (for example a presigned S3 GET URL), same as for video uploads.

<Tip>
  We recommend using the [Developer Portal](https://platform.tavus.io/dev/replicas/create) to upload an image as it provides real-time validation to ensure it meets all requirements before training.
</Tip>

## Image Requirements

Upload a clear, front-facing headshot that meets the following requirements:

* **Formats:** JPG or PNG
* **Minimum resolution:** 512×512 pixels
* **Only one person** visible in the image
* **Head and shoulders** clearly visible in frame
* **No glasses, hats, or face-covering accessories**
* **Avoid visible jewelry** such as large earrings or necklaces
* **Keep hair behind the shoulders** and away from the face and neck
* **Use even lighting** with minimal shadows across the face

<Frame>
  <img src="https://mintcdn.com/tavus/N0ChN_R5w8O9WLNa/images/Screenshot-2026-05-08-060356.png?fit=max&auto=format&n=N0ChN_R5w8O9WLNa&q=85&s=a71a75ffa6fad513a6b178494e1c509e" alt="Screenshot 2026 05 08 060356" width="1788" height="1054" data-path="images/Screenshot-2026-05-08-060356.png" />
</Frame>

Image-based training is a faster and simpler way to create a replica without recording a training video. It offers a simpler setup and is ideal for quick prototyping or AI-generated characters.

<Warning>
  Images will **not** work if they contain multiple people, subjects under 18, non-human characters, visible accessories (such as glasses, headphones, or jewelry), hair in front of shoulders, off-center framing, or unnatural poses such as leaning or lying down.
</Warning>

## AI Image Fixer

If your uploaded image doesn't fully meet the requirements above, set **`auto_fix_training_image`** to `true` when calling [Create Replica](/api-reference/phoenix-replica-model/create-replica). Tavus's AI Image Fixer instantly fixes the uploaded image to fit our requirements, eliminating the need for editing or recapturing photos.

```json theme={null}
{
  "replica_name": "my_image_replica",
  "train_image_url": "https://example.com/headshot.png",
  "voice_name": "anna",
  "auto_fix_training_image": true
}
```

## How `voice_name` works

Image-based training does not create a new voice from your source material. Instead, you must set **`voice_name`** to a **stock voice** identifier slug (for example `anna`). This selects a voice tied to an existing Tavus stock replica so the trained replica has a usable default voice.

<h4 id="example-voice-name-table">
  Example `voice_name` values
</h4>

Below are **example** `voice_name` slugs with a short sample clip for each.

<div className="grid grid-cols-1 gap-4 sm:grid-cols-2 lg:grid-cols-3">
  <div className="rounded-xl border border-zinc-200 p-3 dark:border-zinc-700">
    ```text theme={null}
    benjamin
    ```

    <video controls playsInline preload="metadata" className="max-h-24 w-full max-w-[200px] rounded-lg" src="https://cdn.replica.tavus.io/20269/3448746b_normalized.mp4">
      Your browser does not support the video tag.
    </video>
  </div>

  <div className="rounded-xl border border-zinc-200 p-3 dark:border-zinc-700">
    ```text theme={null}
    james
    ```

    <video controls playsInline preload="metadata" className="max-h-24 w-full max-w-[200px] rounded-lg" src="https://cdn.replica.tavus.io/43019/8a1618dc_normalized.mp4">
      Your browser does not support the video tag.
    </video>
  </div>

  <div className="rounded-xl border border-zinc-200 p-3 dark:border-zinc-700">
    ```text theme={null}
    liam
    ```

    <video controls playsInline preload="metadata" className="max-h-24 w-full max-w-[200px] rounded-lg" src="https://cdn.replica.tavus.io/31783/4c295058_normalized.mp4">
      Your browser does not support the video tag.
    </video>
  </div>

  <div className="rounded-xl border border-zinc-200 p-3 dark:border-zinc-700">
    ```text theme={null}
    anna
    ```

    <video controls playsInline preload="metadata" className="max-h-24 w-full max-w-[200px] rounded-lg" src="https://cdn.replica.tavus.io/40013/7409bd85_normalized.mp4">
      Your browser does not support the video tag.
    </video>
  </div>

  <div className="rounded-xl border border-zinc-200 p-3 dark:border-zinc-700">
    ```text theme={null}
    julia
    ```

    <video controls playsInline preload="metadata" className="max-h-24 w-full max-w-[200px] rounded-lg" src="https://cdn.replica.tavus.io/39359/cd603e65_normalized.mp4">
      Your browser does not support the video tag.
    </video>
  </div>

  <div className="rounded-xl border border-zinc-200 p-3 dark:border-zinc-700">
    ```text theme={null}
    ivy
    ```

    <video controls playsInline preload="metadata" className="max-h-24 w-full max-w-[200px] rounded-lg" src="https://cdn.replica.tavus.io/35249/6198e87b_normalized.mp4">
      Your browser does not support the video tag.
    </video>
  </div>
</div>

<Note>
  When you run **Conversational Video Interface (CVI)** sessions later, you are **not locked** into that stock voice for every conversation. You can attach a [persona](/api-reference/personas/create-persona) whose TTS layer uses an external voice (from Cartesia or ElevenLabs). See [Text-to-Speech (TTS)](/sections/conversational-video-interface/persona/tts) for how to set `external_voice_id` and related fields.
</Note>

### Consent, rights, and acceptable use

By using the image training API, you **affirm that you have the rights** to use the image you supply (for example likeness and publicity rights where applicable). Tavus may **reject** images that appear to depict unauthorized or impermissible subjects.

<Note>
  Replica training typically takes **3–4 hours**.
</Note>