CNN+LSTM Architecture for Speech Emotion Recognition with Data Augmentation
Sound
2018-09-13 v2 Computation and Language
Machine Learning
Audio and Speech Processing
Abstract
In this work we design a neural network for recognizing emotions in speech, using the IEMOCAP dataset. Following the latest advances in audio analysis, we use an architecture involving both convolutional layers, for extracting high-level features from raw spectrograms, and recurrent ones for aggregating long-term dependencies. We examine the techniques of data augmentation with vocal track length perturbation, layer-wise optimizer adjustment, batch normalization of recurrent layers and obtain highly competitive results of 64.5% for weighted accuracy and 61.7% for unweighted accuracy on four emotions.
Cite
@article{arxiv.1802.05630,
title = {CNN+LSTM Architecture for Speech Emotion Recognition with Data Augmentation},
author = {Caroline Etienne and Guillaume Fidanza and Andrei Petrovskii and Laurence Devillers and Benoit Schmauch},
journal= {arXiv preprint arXiv:1802.05630},
year = {2018}
}
Comments
5 pages, 3 figures