Talking Avatar

Beta

Avatar Image

Upload the image for the avatar.

Audio Source

Text for Speech

Lipsync Model

Voice Settings

Voice

Only used when Audio Source is set to Text-to-Speech.

Voice Speed — 1.0×

AI Model

Prompt

Input Image (optional)

Both models support image-to-video (I2V) generation.

Reference Images (Veo only — up to 3)

Ref 1 Ref 2 Ref 3

Only works with 16:9 ratio + 8s duration.

Aspect Ratio

Resolution

Duration — 8s

Veo 3.1 supports 5–8 seconds.

Negative Prompt (optional)

Audio is always disabled — this generates a silent video template ready for lipsync processing.

Ready to Animate

Upload an avatar image and provide audio or text. Your talking avatar video will appear here.