Towards Speech Emotion Recognition "in the wild" using Aggregated Corpora and Deep Multi-Task Learning
This work addresses the problem of generalizing emotion recognition to diverse real-world conditions for applications in human-computer interaction, though it appears incremental as it builds on existing MTL techniques.
The paper tackled the challenge of Speech Emotion Recognition (SER) 'in the wild' by proposing a Multi-Task Learning (MTL) method using gender and naturalness as auxiliary tasks, which significantly improved performance over Single-Task Learning (STL) methods in cross-corpus experiments.
One of the challenges in Speech Emotion Recognition (SER) "in the wild" is the large mismatch between training and test data (e.g. speakers and tasks). In order to improve the generalisation capabilities of the emotion models, we propose to use Multi-Task Learning (MTL) and use gender and naturalness as auxiliary tasks in deep neural networks. This method was evaluated in within-corpus and various cross-corpus classification experiments that simulate conditions "in the wild". In comparison to Single-Task Learning (STL) based state of the art methods, we found that our MTL method proposed improved performance significantly. Particularly, models using both gender and naturalness achieved more gains than those using either gender or naturalness separately. This benefit was also found in the high-level representations of the feature space, obtained from our method proposed, where discriminative emotional clusters could be observed.