Speech to Video

Generate video driven by speech audio input using Wan 2.2.

Overview

The Speech to Video node creates video content driven by speech audio. It uses Wan 2.2 to generate visual content that corresponds to the spoken input, offering fine-grained control over generation parameters.

Configuration

Field Type Default Description
Resolution Select 720p Output resolution: 480p, 580p, 720p
Prompt Textarea Scene description to guide visuals
Negative Prompt Textarea Elements to exclude

Advanced Settings (Collapsible)

Field Type Default Description
Seed Number Random Fixed seed for reproducibility
Num Frames Number Frame count: 16-81
FPS Number Frames per second: 8-24
Inference Steps Number Quality steps: 1-50
Guidance Scale Number Prompt adherence: 0-20
Shift Number Shift parameter: 0-20

Inputs & Outputs

Inputs:

Outputs:

Common Use Cases

Tips