# My lips are concealed: Audio-visual speech enhancement through obstructions

## TL;DR

This paper presents a deep audio-visual speech enhancement model capable of isolating a speaker's voice even with visual occlusions, by leveraging lip movements and voice representations learned through enrollment or self-enrollment.

## Contribution

The introduced model uniquely combines visual and voice cues, maintains performance during visual occlusions, and is trained with artificial mouth occlusions for robustness.

## Key findings

- Improves speech separation during visual occlusion scenarios.
- Learns speaker-specific voice representations on-the-fly.
- Demonstrates effectiveness on unseen speakers.

## Abstract

Our objective is an audio-visual model for separating a single speaker from a mixture of sounds such as other speakers and background noise. Moreover, we wish to hear the speaker even when the visual cues are temporarily absent due to occlusion. To this end we introduce a deep audio-visual speech enhancement network that is able to separate a speaker's voice by conditioning on both the speaker's lip movements and/or a representation of their voice. The voice representation can be obtained by either (i) enrollment, or (ii) by self-enrollment -- learning the representation on-the-fly given sufficient unobstructed visual input. The model is trained by blending audios, and by introducing artificial occlusions around the mouth region that prevent the visual modality from dominating. The method is speaker-independent, and we demonstrate it on real examples of speakers unheard (and unseen) during training. The method also improves over previous models in particular for cases of occlusion in the visual modality.