Transferable Positive/Negative Speech Emotion Recognition via Class-wise Adversarial Domain Adaptation
This work addresses domain adaptation challenges for speech emotion recognition, enabling more robust intelligent agents, but it is incremental as it builds on existing adversarial methods.
The paper tackled the problem of domain shift in speech emotion recognition by proposing class-wise adversarial domain adaptation, which improved performance on positive/negative emotion recognition across different corpora, achieving effective results with limited labeled target data.
Speech emotion recognition plays an important role in building more intelligent and human-like agents. Due to the difficulty of collecting speech emotional data, an increasingly popular solution is leveraging a related and rich source corpus to help address the target corpus. However, domain shift between the corpora poses a serious challenge, making domain shift adaptation difficult to function even on the recognition of positive/negative emotions. In this work, we propose class-wise adversarial domain adaptation to address this challenge by reducing the shift for all classes between different corpora. Experiments on the well-known corpora EMODB and Aibo demonstrate that our method is effective even when only a very limited number of target labeled examples are provided.