Domain Adversarial Learning For Emotion Recognition

Abstract

In practical applications for emotion recognition, users do not always exist in the training corpus. The mismatch between training speakers and testing speakers affects the performance of the trained model. To deal with this problem, we need our model to focus on emotion-related information, while ignoring the difference between speaker identities. In this paper, we look into the use of the domain adversarial neural network (DANN) to extract a common representation between different speakers. The primary task is to predict emotion labels. The secondary task is to learn a common representation where speaker identities can not be distinguished. By using the gradient reversal layer, the gradients coming from the secondary task are used to bring the representations for different speakers closer. To verify the effectiveness of the proposed method, we conduct experiments on the IEMOCAP database. Experimental results demonstrate that the proposed framework shows an absolute improvement of 3.4

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keylian2019domain

Related papers

• Unsupervised Cross-lingual Speech Emotion Recognition Using Domainadversarial Neural Network (2020)0.00
• Transferable Positive/negative Speech Emotion Recognition Via Class-wise Adversarial Domain Adaptation (2018)9.23
• Adversarial Learning Of Raw Speech Features For Domain Invariant Speech Recognition (2018)9.23
• Towards Adversarial Learning Of Speaker-invariant Representation For Speech Emotion Recognition (2019)0.00
• Adversarial Training For Multi-domain Speaker Recognition (2020)6.77
• Self Supervised Adversarial Domain Adaptation For Cross-corpus And Cross-language Speech Emotion Recognition (2022)13.11
• Improving Speaker-independent Speech Emotion Recognition Using Dynamic Joint Distribution Adaptation (2024)7.50
• Acted Vs. Improvised: Domain Adaptation For Elicitation Approaches In Audio-visual Emotion Recognition (2021)0.00