Searching for Effective Preprocessing Method and CNN-based Architecture with Efficient Channel Attention on Speech Emotion Recognition

TL;DR

This paper introduces an effective preprocessing strategy and a CNN model with efficient channel attention for speech emotion recognition, achieving state-of-the-art results despite limited training data.

Contribution

It proposes a novel combination of multi-resolution preprocessing and ECA-augmented CNN architecture to improve SER performance with fewer parameters.

Findings

Increasing frequency resolution improves emotion recognition accuracy.

ECA blocks enhance channel feature representation efficiently.

Data augmentation with multiple preprocessing methods boosts performance.

Abstract

Speech emotion recognition (SER) classifies human emotions in speech with a computer model. Recently, performance in SER has steadily increased as deep learning techniques have adapted. However, unlike many domains that use speech data, data for training in the SER model is insufficient. This causes overfitting of training of the neural network, resulting in performance degradation. In fact, successful emotion recognition requires an effective preprocessing method and a model structure that efficiently uses the number of weight parameters. In this study, we propose using eight dataset versions with different frequency-time resolutions to search for an effective emotional speech preprocessing method. We propose a 6-layer convolutional neural network (CNN) model with efficient channel attention (ECA) to pursue an efficient model structure. In particular, the well-positioned ECA blocks can improve channel feature representation with only a few parameters. With the interactive emotional dyadic motion capture (IEMOCAP) dataset, increasing the frequency resolution in preprocessing emotional speech can improve emotion recognition performance. Also, ECA after the deep convolution layer can effectively increase channel feature representation. Consequently, the best result (79.37UA 79.68WA) can be obtained, exceeding the performance of previous SER models. Furthermore, to compensate for the lack of emotional speech data, we experiment with multiple preprocessing data methods that augment trainable data preprocessed with all different settings from one sample. In the experiment, we can achieve the highest result (80.28UA 80.46WA).