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

arXiv: https://arxiv.org/abs/1708.07050 · HTML: https://mlsanity.com/paper/1708.07050/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 23 Aug 2017.

## Built on by (most cited first, 6 of 6)

- [Learning Speech Emotion Representations in the Quaternion Domain](https://mlsanity.com/paper/2204.02385/) (2022)
- [EmotionNAS: Two-stream Neural Architecture Search for Speech Emotion Recognition](https://mlsanity.com/paper/2203.13617/) (2022)
- [Best Practices for Noise-Based Augmentation to Improve the Performance of Deployable Speech-Based Emotion Recognition Systems](https://mlsanity.com/paper/2104.08806/) (2021)
- [Accounting for Variations in Speech Emotion Recognition with Nonparametric Hierarchical Neural Network](https://mlsanity.com/paper/2109.04316/) (2021)
- [Dynamic Time-Alignment of Dimensional Annotations of Emotion using Recurrent Neural Networks](https://mlsanity.com/paper/2209.10223/) (2022)
- [Joint Learning using Mixture-of-Expert-Based Representation for Speech Enhancement and Robust Emotion Recognition](https://mlsanity.com/paper/2509.08470/) (2025)

## Abstract (arXiv)

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.