Neural Voice Puppetry: Audio-driven Facial Reenactment

TL;DR

Neural Voice Puppetry is a deep learning method that synthesizes realistic facial videos synchronized with audio inputs, enabling versatile applications like avatars, dubbing, and talking head generation.

Contribution

It introduces a novel neural network approach utilizing a 3D face model for stable, realistic audio-driven facial reenactment that generalizes across different individuals.

Findings

Achieves high realism in synthesized videos.

Demonstrates generalization to unseen speakers.

Outperforms state-of-the-art methods in user studies.

Abstract

We present Neural Voice Puppetry, a novel approach for audio-driven facial video synthesis. Given an audio sequence of a source person or digital assistant, we generate a photo-realistic output video of a target person that is in sync with the audio of the source input. This audio-driven facial reenactment is driven by a deep neural network that employs a latent 3D face model space. Through the underlying 3D representation, the model inherently learns temporal stability while we leverage neural rendering to generate photo-realistic output frames. Our approach generalizes across different people, allowing us to synthesize videos of a target actor with the voice of any unknown source actor or even synthetic voices that can be generated utilizing standard text-to-speech approaches. Neural Voice Puppetry has a variety of use-cases, including audio-driven video avatars, video dubbing, and text-driven video synthesis of a talking head. We demonstrate the capabilities of our method in a series of audio- and text-based puppetry examples, including comparisons to state-of-the-art techniques and a user study.