TRNet: Two-level Refinement Network leveraging Speech Enhancement for Noise Robust Speech Emotion Recognition
This addresses noise robustness in SER, an incremental improvement for practical applications in noisy settings.
The paper tackles the problem of environmental noise degrading Speech Emotion Recognition (SER) performance by introducing TRNet, a two-level refinement network that uses speech enhancement and reference signals to improve robustness, achieving substantial gains in matched and unmatched noisy environments without harming noise-free performance.
One persistent challenge in Speech Emotion Recognition (SER) is the ubiquitous environmental noise, which frequently results in deteriorating SER performance in practice. In this paper, we introduce a Two-level Refinement Network, dubbed TRNet, to address this challenge. Specifically, a pre-trained speech enhancement module is employed for front-end noise reduction and noise level estimation. Later, we utilize clean speech spectrograms and their corresponding deep representations as reference signals to refine the spectrogram distortion and representation shift of enhanced speech during model training. Experimental results validate that the proposed TRNet substantially promotes the robustness of the proposed system in both matched and unmatched noisy environments, without compromising its performance in noise-free environments.