KLING • Avatar / Talking Face

KLING Avatar / Talking Face

Upload one portrait image and one audio file to generate a talking avatar clip with synchronized speech animation.

Model: kwaivgi/kling-avatar-v2

Create Talking Avatar

Best results usually come from a clean front-facing portrait and clear speech audio. :contentReference[oaicite:1]{index=1}

Avatar Image

Use a portrait with visible facial features. :contentReference[oaicite:2]{index=2}

Audio File

Clear recorded speech works best for lip sync. :contentReference[oaicite:3]{index=3}

Image Preview

No image selected

Audio Preview

No audio selected

Prompt

Optional style and emotion guidance. The audio remains the main driver for speech timing. :contentReference[oaicite:4]{index=4}

Negative Prompt

Use this to reduce distortions, flicker, extra mouth shapes, artifacts and text overlays.

Aspect Ratio

Choose the output frame shape.

CFG Scale

Higher value = stronger adherence to prompt.

Advanced JSON Input

Optional. This JSON is merged into the model input so you can test extra model fields without changing the PHP code.

Latest saved records from jobs/kling-avatar-history.json

No avatar jobs yet.