Character-AI Ovi

Ovi (I2V Version)

Ovi is a veo-3 like, image-to-audio-video (I2AV) generation model that creates synchronized video and audio from a single image plus a descriptive text prompt.

It is designed for short-form storytelling, where a still image is brought to life with cinematic motion, dialogue, and sound.

🌟 Key Features

• 🎬 Image → Video+Audio – Bring a static image to life with synchronized audiovisual output.
• 📝 Prompt-driven – Use text prompts to control scene dynamics, style, and audio.
• 🗣️ Speech & Sound – Insert dialogue or sound effects using special tags.
• ⏱️ Short-form Output – Generates 5-second clips at 24 FPS.

💲 Pricing

Billing Rules

• Minimum charge: 5 seconds

🎨 How to Use

1. Upload Image

• Provide a reference image as the base frame.
• Make sure the URL is valid and accessible (a preview should appear).

1. Enter Prompt

• Describe scene motion, style, and atmosphere.

• Use tags for sound:

• <S>... <E> → Speech (converted into spoken audio)

• <AUDCAP>... <ENDAUDCAP> → Background audio / effects

1. Set Seed

• -1 = random output
• Any fixed number = reproducible results

1. Run

• Click Run $0.15 to generate your 5s image-to-audio-video clip.
• Preview and download the result.

📝 Prompt Example

A wide shot of a medieval knight standing in the rain, sword planted into the ground, glowing with mystical energy. 
<S>I will defend this land until my last breath.<E> 
<AUDCAP>Thunder rolls across the dark sky, distant war drums echo.<ENDAUDCAP>

🙏 Acknowledgements

• Wan2.2 – Video backbone initialization
• MMAudio – Audio encoder/decoder inspiration

⭐ Citation

If Ovi is useful, please ⭐ the repo and cite the paper:

@misc{low2025ovitwinbackbonecrossmodal,
 title={Ovi: Twin Backbone Cross-Modal Fusion for Audio-Video Generation}, 
 author={Chetwin Low and Weimin Wang and Calder Katyal},
 year={2025},
 eprint={2510.01284},
 archivePrefix={arXiv},
 primaryClass={cs.MM},
 url={https://arxiv.org/abs/2510.01284}, 
}