EMO: Emote Portrait Alive -- Generating Expressive Portrait Videos with Audio2Video Diffusion Model under Weak Conditions

TL;DR

This paper introduces EMO, a new audio-to-video diffusion model that generates highly expressive and realistic portrait videos directly from audio, overcoming limitations of traditional methods and capturing nuanced facial expressions.

Contribution

EMO is a novel direct audio-to-video synthesis framework that enhances realism and expressiveness without relying on intermediate facial models or landmarks.

Findings

Produces convincing speaking and singing videos

Outperforms state-of-the-art methods in realism and expressiveness

Ensures seamless transitions and identity preservation

Abstract

In this work, we tackle the challenge of enhancing the realism and expressiveness in talking head video generation by focusing on the dynamic and nuanced relationship between audio cues and facial movements. We identify the limitations of traditional techniques that often fail to capture the full spectrum of human expressions and the uniqueness of individual facial styles. To address these issues, we propose EMO, a novel framework that utilizes a direct audio-to-video synthesis approach, bypassing the need for intermediate 3D models or facial landmarks. Our method ensures seamless frame transitions and consistent identity preservation throughout the video, resulting in highly expressive and lifelike animations. Experimental results demonsrate that EMO is able to produce not only convincing speaking videos but also singing videos in various styles, significantly outperforming existing state-of-the-art methodologies in terms of expressiveness and realism.