Skip to main content
Character-3 is Hedra’s proprietary audio-to-video model that transforms static images into expressive video content with natural character movements.

​
How It Works

1

Add an Image

Upload or generate an image using models like Imagen4, Flux Kontext, or others.
2

Create Audio

Generate audio with ElevenLabs or use your own audio clip.
3

Generate Video

The system produces finished videos in just a few minutes.

​
Use Cases

Interviews

Create realistic interview-style content

Tutorials

Generate educational explainer videos

Animated Shorts

Produce short-form animated content

Character Animation

Bring characters to life with natural movement

​
Multi-Character Videos

Character-3 supports selecting which character speaks in multi-character images:
1

Upload Image

Upload or create your starting image containing multiple characters.
2

Select Speaker

Click on the uploaded image and click the “Select Speaker” button that appears when you hover over the image.
3

Mark the Character

Position the green smiley icon on your chosen character by dragging and expanding it. This character will animate according to your audio input.
Once you’ve marked a character using the green smiley indicator, proceed with video creation and that character will be the one performing the audio you provide.