Focal Loss based Residual Convolutional Neural Network for Speech Emotion Recognition

2019/06/11 by Tripathi, Suraj, Kumar, Abhay, Ramesh, Abhiram +2
#Artificial Intelligence (cs.AI) #Audio and Speech Processing (eess.AS) #FOS: Computer and information sciences #FOS: Electrical engineering #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Sound (cs.SD) #electronic engineering #information engineering

paper · doi:10.48550/arxiv.1906.05682

Abstract

This paper proposes a Residual Convolutional Neural Network (ResNet) based on speech features and trained under Focal Loss to recognize emotion in speech. Speech features such as Spectrogram and Mel-frequency Cepstral Coefficients (MFCCs) have shown the ability to characterize emotion better than just plain text. Further Focal Loss, first used in One-Stage Object Detectors, has shown the ability to focus the training process more towards hard-examples and down-weight the loss assigned to well-classified examples, thus preventing the model from being overwhelmed by easily classifiable examples.

Related

• Learning Discriminative features using Center Loss and Reconstruction as Regularizer for Speech Emotion Recognition
• End-to-end Triplet Loss based Emotion Embedding System for Speech Emotion Recognition
• Emotion Recognition From Speech With Recurrent Neural Networks
• Deep Learning for Speech Emotion Recognition: A CNN Approach Utilizing Mel Spectrograms
• Deep Learning based Emotion Recognition System Using Speech Features and Transcriptions