Capturing Long-term Temporal Dependencies with Convolutional Networks for Continuous Emotion Recognition

TL;DR

This paper introduces a novel downsampling and upsampling convolutional network architecture that effectively captures long-term temporal dependencies for continuous emotion recognition, producing smooth and accurate emotion trajectories.

Contribution

The paper proposes a downsampling/upsampling convolutional network to better model long-term dependencies and generate smooth emotion trajectories, outperforming previous methods on RECOLA.

Findings

Achieves state-of-the-art audio-only performance on RECOLA

Produces smooth and consistent emotion trajectories

Outperforms previous architectures using dilated convolutions

Abstract

The goal of continuous emotion recognition is to assign an emotion value to every frame in a sequence of acoustic features. We show that incorporating long-term temporal dependencies is critical for continuous emotion recognition tasks. To this end, we first investigate architectures that use dilated convolutions. We show that even though such architectures outperform previously reported systems, the output signals produced from such architectures undergo erratic changes between consecutive time steps. This is inconsistent with the slow moving ground-truth emotion labels that are obtained from human annotators. To deal with this problem, we model a downsampled version of the input signal and then generate the output signal through upsampling. Not only does the resulting downsampling/upsampling network achieve good performance, it also generates smooth output trajectories. Our method yields the best known audio-only performance on the RECOLA dataset.