Enhancing Speech Emotion Recognition Through Differentiable Architecture Search
This work addresses the challenge of automating architecture design for SER, which is critical for emotion-aware human-computer interaction, but it is incremental as it builds on existing DARTS methods with a specific novel mechanism.
The paper tackles the problem of designing optimal deep learning architectures for Speech Emotion Recognition (SER) by proposing a Differentiable Architecture Search (DARTS)-optimized joint CNN and LSTM architecture, achieving significantly higher SER accuracy than hand-engineered CNN-LSTM configurations and outperforming previous DARTS-based SER results on the IEMOCAP and MSP-IMPROV datasets.
Speech Emotion Recognition (SER) is a critical enabler of emotion-aware communication in human-computer interactions. Recent advancements in Deep Learning (DL) have substantially enhanced the performance of SER models through increased model complexity. However, designing optimal DL architectures requires prior experience and experimental evaluations. Encouragingly, Neural Architecture Search (NAS) offers a promising avenue to determine an optimal DL model automatically. In particular, Differentiable Architecture Search (DARTS) is an efficient method of using NAS to search for optimised models. This paper proposes a DARTS-optimised joint CNN and LSTM architecture, to improve SER performance, where the literature informs the selection of CNN and LSTM coupling to offer improved performance. While DARTS has previously been applied to CNN and LSTM combinations, our approach introduces a novel mechanism, particularly in selecting CNN operations using DARTS. In contrast to previous studies, we refrain from imposing constraints on the order of the layers for the CNN within the DARTS cell; instead, we allow DARTS to determine the optimal layer order autonomously. Experimenting with the IEMOCAP and MSP-IMPROV datasets, we demonstrate that our proposed methodology achieves significantly higher SER accuracy than hand-engineering the CNN-LSTM configuration. It also outperforms the best-reported SER results achieved using DARTS on CNN-LSTM.