Unsupervised Adversarial Domain Adaptation for Cross-Lingual Speech Emotion Recognition

• Published in: International Conference on Affective Computing and Intelligent Interaction and workshops pp. 732 - 737
• Main Authors: Latif, Siddique, Qadir, Junaid, Bilal, Muhammad
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.09.2019
• Subjects: Adaptation models; Data models; domain adaptation; Emotion recognition; Gallium nitride; generative adversarial networks (GANs); Multi-lingual; Robustness; Speech emotion recognition; Task analysis; Training; Urdu language

Cross-lingual speech emotion recognition (SER)is a crucial task for many real-world applications. The performance of SER systems is often degraded by the differences in the distributions of training and test data. These differences become more apparent when training and test data belong to different...