A Dual-Staged Context Aggregation Method Towards Efficient End-To-End Speech Enhancement
This work addresses speech enhancement for noisy audio processing, offering an incremental improvement in efficiency and performance.
The paper tackled the challenge of efficiently aggregating contextual information in end-to-end speech enhancement by proposing a densely connected convolutional and recurrent network (DCCRN), which improved STOI by 0.23 and PESQ by 1.38 on average compared to baselines.
In speech enhancement, an end-to-end deep neural network converts a noisy speech signal to a clean speech directly in time domain without time-frequency transformation or mask estimation. However, aggregating contextual information from a high-resolution time domain signal with an affordable model complexity still remains challenging. In this paper, we propose a densely connected convolutional and recurrent network (DCCRN), a hybrid architecture, to enable dual-staged temporal context aggregation. With the dense connectivity and cross-component identical shortcut, DCCRN consistently outperforms competing convolutional baselines with an average STOI improvement of 0.23 and PESQ of 1.38 at three SNR levels. The proposed method is computationally efficient with only 1.38 million parameters. The generalizability performance on the unseen noise types is still decent considering its low complexity, although it is relatively weaker comparing to Wave-U-Net with 7.25 times more parameters.