Knowledge Transfer For On-Device Speech Emotion Recognition with Neural Structured Learning
This work addresses the problem of enabling efficient SER on edge devices for human-computer interaction applications, representing an incremental improvement in transfer learning for resource-constrained settings.
The paper tackles the challenge of deploying speech emotion recognition (SER) on edge devices with limited memory and computational resources by proposing a neural structured learning framework that uses synthesized graphs for knowledge transfer, resulting in lightweight models with enhanced performance compared to baseline methods.
Speech emotion recognition (SER) has been a popular research topic in human-computer interaction (HCI). As edge devices are rapidly springing up, applying SER to edge devices is promising for a huge number of HCI applications. Although deep learning has been investigated to improve the performance of SER by training complex models, the memory space and computational capability of edge devices represents a constraint for embedding deep learning models. We propose a neural structured learning (NSL) framework through building synthesized graphs. An SER model is trained on a source dataset and used to build graphs on a target dataset. A relatively lightweight model is then trained with the speech samples and graphs together as the input. Our experiments demonstrate that training a lightweight SER model on the target dataset with speech samples and graphs can not only produce small SER models, but also enhance the model performance compared to models with speech samples only and those using classic transfer learning strategies.