Omnihuman

By: ByteDance Research Video creation

Image-to-video lip-sync generation with ByteDance Omnihuman v1.5.

Description

ByteDance Omnihuman v1.5 produces expressive talking-head videos from a single image paired with an audio track in multiple languages. Supply both, or let us craft the narration for you:

• 🎤 **Multilingual & Flexible Audio:** Upload custom audio or generate fresh narration with ElevenLabs-quality speech in languages like English, French, Spanish, German, and more.
• 🧠 **Smart Scripting:** Describe the tone and topic, and we’ll write the script in your desired language when you don’t have one ready.
• 💋 **Precision Lip-Sync:** Facial expressions and mouth movements track the audio closely, creating a natural and believable performance.

Perfect for creating localized product updates, global marketing teasers, and rapid avatar content production.

What can this MCP do?

Generate Omnihuman video

Create a lip-synced video from an image and audio, with optional audio synthesis.

Example prompts:

Animate this headshot with a 20-second welcome script in French, using a female narrator.

Use this product photo and the attached podcast intro audio to create a 15-second promo clip.

Generate calm narration in German introducing Simtheory analytics, male voice, 10 seconds, friendly tone.

Lip-sync this uploaded MP3 with the CEO portrait for our investor update.

Write and speak a 25-second inspirational quote in Spanish using my supplied image.

Copy these prompts to get started quickly with Omnihuman in your Simtheory workspace.

How It Works

1. Install the MCP

Install Omnihuman from your Simtheory workspace with one click.

2. Connect Your Account

Securely authorize access to your Omnihuman account.

3. Start Using

Your AI can now interact with Omnihuman on your behalf.

Users also installed these MCPs

Video maker

Generate videos from a topic or a provided script with narrated voiceover, captions (optional) and AI B‑roll.

Lucy Edit Pro

Edit videos from a text prompt using Lucy Edit Pro (decart/lucy-edit/pro).

Sora 2

Generate cinematic videos from text or images with OpenAI Sora 2. Choose aspect ratio (16:9 landscape or 9:16 portrait), set size (1280x720 or 720x1280), and select length (4s, 8s, or 12s). Includes synced audio by default.

Decart Image-to-Video

Animate a single image into a smooth 720p video using Decart Lucy-14B.

WAN 2.2 Media Suite

Generate high-quality images and videos from text or images using WAN 2.2. Features a standard model for detailed control and a turbo model for ultra-fast video generation.

Veo3.1

Generate high-fidelity 720p/1080p videos from text, images, or reference photos using Google Veo3.1.

Ready to use Omnihuman MCP Server?

Get Started with Simtheory