Collaborative Learning to Generate Audio-Video Jointly

TL;DR

This paper introduces a novel method for jointly generating realistic audio and video data using GANs, addressing the previously underexplored area of multi-modal multimedia synthesis.

Contribution

It proposes a multi-discriminator GAN framework for correlated audio-video generation and provides a new dataset for this task.

Findings

Generated realistic audio-video samples validated by standard metrics

Achieved high Inception Score and low FID scores

Human evaluation confirms the naturalness of generated data

Abstract

There have been a number of techniques that have demonstrated the generation of multimedia data for one modality at a time using GANs, such as the ability to generate images, videos, and audio. However, so far, the task of multi-modal generation of data, specifically for audio and videos both, has not been sufficiently well-explored. Towards this, we propose a method that demonstrates that we are able to generate naturalistic samples of video and audio data by the joint correlated generation of audio and video modalities. The proposed method uses multiple discriminators to ensure that the audio, video, and the joint output are also indistinguishable from real-world samples. We present a dataset for this task and show that we are able to generate realistic samples. This method is validated using various standard metrics such as Inception Score, Frechet Inception Distance (FID) and through human evaluation.