Reference image
The reference image is the foundation of how your LemonSlice avatar looks and talks. Our model picks up on every detail of the image and uses it to infer micro-expressions, poses, and even background motion. Follow these tips for best avatar quality.
Iterate on your images in the LemonSlice web app with our built-in AI image editor.
Larger faces have best facial details. Make the head as large as possible. If you want your avatar to use hand gestures, use a half-body portrait while maintaining a relatively large face.
Don’t zoom out too much. Our model renders every pixel in real time, so focus those pixels on the character, not the background.
Use a neutral expression and look directly into the camera. Your reference image controls the expression and pose of your avatar throughout the call. For example, a reference image of a person looking up may cause the avatar to always be looking up.
For non-human or cartoon characters, use an image with the mouth slightly open. This will show the model how you want the mouth to animate.
For humans: Use this approach if you’re noticing degradation on the quality of the teeth.
We’ve found that running your image through this image edit prompt does this the best:The exact same image, but they are talking (with their mouth open).
Drive simple animations from a single image. Our model intelligently animates simple scenes using the reference image, the agent prompt, and the idle prompt. Keep your animations scoped to relatively simple, looping animations like waves in water and atmospheric effects. The model can also animate people in the background, but it works best when they’re out of focus.
When holding objects, use one hand in a natural pose. This keeps the avatar stable and expressive, maintains clean motion, and avoids awkward or unnatural positioning that can make the interaction feel off.
Choosing a voice
Voice intonation impacts facial expression. For example, choosing a voice with an angry intonation will yield an angry expression. Likewise, a monotonous voice will produce a neutral facial expression.
If your avatar’s mouth looks off, choose another voice. Some voices work better than others.