Discriminative Feature Representation Based on Cascaded Attention Network with Adversarial Joint Loss for Speech Emotion Recognition

Liu, Yang; Sun, Haoqin; Guan, Wenbo; Xia, Y.; Zhang, Zhao

doi:10.21437/interspeech.2022-11480

Cited by 5 publications

(1 citation statement)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, one challenge for SER and Emotional TTS is the lack of data, which has not been taken into account in most related approaches [3,4,5,6,7]. In reality, emotional speech data is more difficult and expensive to acquire compared to Neutral style speech, which results in a common issue that the speech data distributions are highly skewed toward the Neutral emotion class.…”

Section: Introductionmentioning

confidence: 99%

Generative Data Augmentation Guided by Triplet Loss for Speech Emotion Recognition

Wang¹,

Hemati²,

Guðnason³

et al. 2022

Interspeech 2022

View full text Add to dashboard Cite

Speech Emotion Recognition (SER) is crucial for humancomputer interaction but still remains a challenging problem because of two major obstacles: data scarcity and imbalance. Many datasets for SER are substantially imbalanced, where data utterances of one class (most often Neutral) are much more frequent than those of other classes. Furthermore, only a few data resources are available for many existing spoken languages. To address these problems, we exploit a GAN-based augmentation model guided by a triplet network, to improve SER performance given imbalanced and insufficient training data. We conduct experiments and demonstrate: 1) With a highly imbalanced dataset, our augmentation strategy significantly improves the SER performance (+8% recall score compared with the baseline). 2) Moreover, in a cross-lingual benchmark, where we train a model with enough source language utterances but very few target language utterances (around 50 in our experiments), our augmentation strategy brings benefits for the SER performance of all three target languages.

show abstract

Section: Introductionmentioning

confidence: 99%