Create a talking avatar with Hedra

Step-by-step

1. Click on Additional Tools from the homepage and select Animate an Avatar
2. Minimize the sidebar to view the Generations and Templates sections
3. Click Upload in the Start Frame section to select an avatar image from your device
4. Scroll to the bottom to select a model, such as Hedra Character 3
5. Set the Duration to Auto to match the video length to the audio
6. Select your preferred Aspect Ratio and Resolution
7. Click Upload in the Add Speech section to add your audio file
8. Click the Generate button and wait for the queue to complete
9. Download the finished video from the Generations section

Options

• Generate an image, take a photo, or use a library image as a start frame
• Generate audio via 11 Labs/Minimax, record directly, or use library audio
• Trim audio or remove background noise using built-in tools
• Swap voices using 11 Labs or Minimax integrations
• Add a text prompt to influence the generation
• Choose between Hedra Avatar, Hedra Omnia, Cling, or Omnihuman models

Watch out for

• Generation time depends on current server traffic and your specific subscription tier
• Higher tier plans receive priority in the generation queue
• Credit usage varies based on the specific model selected and the length of the audio

Tips

• Check the short description and credits-per-second for each model to decide which fits your project
• Leave resolution settings at their defaults for standard results
• Bypass the prompt feature if you prefer manual control, as it may not always yield desired results
• Use 11 Labs for voice selection as it is a reliable industry standard

Highlights

“With just an image and audio, you can create realistic talking avatars in minutes.”