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

# OmniHuman

> ByteDance's human-animation model in ImagineArt Lipsync Studio — create a talking video from one image and audio.

## OmniHuman

OmniHuman is a human video generation model from ByteDance. It brings a single image to life from an audio signal, generating speech, expressions and natural body movement. Per the [OmniHuman project page](https://omnihuman-lab.github.io/), it handles portrait, half-body and full-body images. In ImagineArt it appears in **Lipsync Studio** as **OmniHuman - Bytedance**, described as "Realistic talking Avatars".

## Specifications

| Feature | Details |
| - | - |
| **Developer** | ByteDance |
| **Where to find it** | Video Studio → **Lipsync Studio** |
| **Inputs** | One image (JPG, JPEG, PNG or WEBP, up to 30MB), audio and an optional prompt |
| **Audio** | Upload an audio file with **Add Audio**, or create one with **Generate Audio** |
| **Duration** | Set by the length of your audio |
| **Settings** | No resolution, duration or aspect ratio controls |

## How to use

<Steps>
  <Step title="Open Lipsync Studio">
    Go to **Video** and select **Lipsync Studio** from the Video Tools tabs.
  </Step>

  <Step title="Select the model">
    Open **Select Model** and choose **OmniHuman - Bytedance**.
  </Step>

  <Step title="Add your image, audio and prompt">
    Upload the image under **Upload Image**, then click **Add Audio** or **Generate Audio**. Use **Prompt** to describe gestures or mood.
  </Step>

  <Step title="Generate">
    Click **Generate**.
  </Step>
</Steps>

## Other Lipsync Studio models

| Model | Resolution | Inputs |
| - | - | - |
| **OmniHuman** | — | Image + audio + prompt |
| [Fabric 1.0](/ai-models/video/fabric-1-0) | 480p–720p | Image + audio |
| [Infini Talk](/ai-models/video/infini-talk) | 480p–720p | Image + audio + prompt |
| [Kling Avatars 2.0 Pro](/ai-models/video/kling-avatars-2-0-pro) | — | Image + audio + prompt |
| [Wan 2.5 Speak](/ai-models/video/wan-2-5-speak) | 480p–1080p | Image + audio text + prompt |


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.