ToonMageV2 generates high fidelity facial images
Audio Conditioned LipSync with Latent Diffusion Models