Multi-Microphone Speech Emotion Recognition using the Hierarchical Token-semantic Audio Transformer Architecture
This addresses robustness issues for emotion recognition systems in adverse acoustic environments, but it is incremental as it adapts an existing transformer model to multi-channel inputs.
The study tackled performance degradation of speech emotion recognition in real-life reverberant conditions by processing multi-microphone signals, achieving superior accuracy compared to single-channel baselines.
The performance of most emotion recognition systems degrades in real-life situations ('in the wild' scenarios) where the audio is contaminated by reverberation. Our study explores new methods to alleviate the performance degradation of SER algorithms and develop a more robust system for adverse conditions. We propose processing multi-microphone signals to address these challenges and improve emotion classification accuracy. We adopt a state-of-the-art transformer model, the HTS-AT, to handle multi-channel audio inputs. We evaluate two strategies: averaging mel-spectrograms across channels and summing patch-embedded representations. Our multi-microphone model achieves superior performance compared to single-channel baselines when tested on real-world reverberant environments.