Before You Start
You will need the following:- Image (uploaded or generated in Hedra)
- Audio (uploaded, recorded, or AI-generated)
- Access to video generation inside Hedra Studio
See: Create Images
Tip: Front-facing portraits produce the most natural speaking results.
See: Create Voice
Tip: Clear audio improves lip-sync accuracy.
Select an Avatar Video Model
- Go to the Avatar section
- Choose an avatar video model (Hedra Omnia, Avatar, Kling) from the pop-up menu below the prompt box
- Select aspect ratio, resolution and batch size
- Confirm image and audio are both attached loaded
Generate the Video
- In the prompt box, describe your video
- Click to Generate
- Wait for processing to complete
- Preview the result
- On the video, select to:
View Details, Share, Download or Delete if needed
Best Practices
- Use high-resolution images
- Avoid extreme angles or heavy shadows
- Keep audio clean and evenly paced
- Start with short clips when testing
Troubleshooting
The lips don’t match the audio- Check for background noise
- Ensure the audio file is clear and not distorted
The face looks unnatural- Use a well-lit, front-facing image
- Avoid heavy filters or stylization
The video failed to generate- Confirm both image and audio are attached
- Try regenerating
Next Steps
- Enhance motion with motion control tools
- Experiment with different voice styles
- Create multi-scene videos using Composer