# 2.5D Visual Sound

## TL;DR

This paper introduces a deep learning method that converts monaural audio into binaural, spatialized sound by utilizing visual cues from video, enabling richer 3D audio experiences without specialized equipment.

## Contribution

It presents a novel multi-modal neural network that leverages visual information to generate binaural audio from monaural recordings in a self-supervised manner.

## Key findings

- Effective conversion of monaural to binaural audio using visual cues
- Improved audio-visual source separation performance
- Self-supervised learning benefits for spatial audio generation

## Abstract

Binaural audio provides a listener with 3D sound sensation, allowing a rich perceptual experience of the scene. However, binaural recordings are scarcely available and require nontrivial expertise and equipment to obtain. We propose to convert common monaural audio into binaural audio by leveraging video. The key idea is that visual frames reveal significant spatial cues that, while explicitly lacking in the accompanying single-channel audio, are strongly linked to it. Our multi-modal approach recovers this link from unlabeled video. We devise a deep convolutional neural network that learns to decode the monaural (single-channel) soundtrack into its binaural counterpart by injecting visual information about object and scene configurations. We call the resulting output 2.5D visual sound---the visual stream helps "lift" the flat single channel audio into spatialized sound. In addition to sound generation, we show the self-supervised representation learned by our network benefits audio-visual source separation. Our video results: http://vision.cs.utexas.edu/projects/2.5D_visual_sound/