# Lipsync

### Summary

The Lipsync Node uses AI models to generate a video of a character speaking, starting from a single image. Provide a character image, write a prompt or connect an audio track, and the AI produces a realistic video with natural lip movements, facial expressions, and synchronized speech. Models can generate native audio directly from your prompt, or you can connect your own audio for the character to speak to.

It works like a video generation model with built-in speaking capabilities turning any portrait or character into a talking, expressive video.

***

### How to Use

1. Add the Node:
   * Click the Add (+) button and select Lipsync from the Video node category.
2. Connect a Character Image:
   * Link a portrait, headshot, or character image via the Character Image input handle. This is the face the AI will animate.
3. Provide Speech Input:
   * With native audio: Write a prompt describing what the character should say, and the AI generates both the video and the spoken audio.
   * With connected audio: Connect an external audio track, and the AI generates lip movements and expressions matched to that audio.
4. Configure Settings:
   * Select your preferred Model, Duration, Resolution, and Aspect Ratio from the Properties panel.
5. Generate:
   * Click Run, and the AI will produce a video of your character speaking with synchronized lip movements.

***

### Choosing the Right Settings

| Setting        | Type                             | Impact on Output                                                                                                                             |
| -------------- | -------------------------------- | -------------------------------------------------------------------------------------------------------------------------------------------- |
| Model          | Dropdown                         | Selects the AI model used for generation. Different models vary in speech realism, facial expression quality, and native audio capabilities. |
| Duration       | Dropdown (e.g., 5s, 10s)         | Sets the length of the generated video.                                                                                                      |
| Resolution     | Dropdown (e.g., 720p, 1080p)     | Determines the output video resolution. Higher resolutions produce sharper facial details.                                                   |
| Aspect Ratio   | Dropdown (16:9, 9:16, 1:1, etc.) | Defines the video frame dimensions.                                                                                                          |
| Generate Audio | Checkbox                         | When enabled, the AI generates spoken audio natively from your prompt. Disable if you're connecting your own audio track.                    |
| Seed           | Number Input                     | A fixed number for reproducible results across generations.                                                                                  |

***

### Sample Use Cases

<details open>

<summary>AI Influencer Content</summary>

Generate a consistent AI character that delivers daily content—product reviews, trend commentary, or motivational clips—without ever stepping in front of a camera. One portrait, endless talking videos.

</details>

<details open>

<summary>Personalized Video Outreach at Scale</summary>

Turn a single headshot into hundreds of personalized sales or onboarding videos. Connect different audio scripts for each prospect or customer, and batch-generate tailored talking-head messages.

</details>

<details open>

<summary>UGC-Style Ads Without Creators</summary>

Produce authentic-looking user-generated-content-style video ads from character images. Skip the creator booking and turnaround time—generate multiple variations with different scripts and languages instantly.

</details>

***

### Lipsync Models

Visit [Video Models](https://help.imagine.art/workflows/understanding-nodes/video-nodes/video-models) to explore all available models and find the one that fits your needs for creating or transforming videos.