CheapNET: Improving Light-weight speech enhancement network by projected loss function
This work addresses efficient speech enhancement for smart devices and real-time communication, offering incremental improvements in performance.
The paper tackled noise suppression and echo cancellation in speech enhancement by introducing a novel projection loss function, achieving near state-of-the-art results with 3.1M parameters and 0.4GFlops/s, and outperforming industry-leading models in echo cancellation.
Noise suppression and echo cancellation are critical in speech enhancement and essential for smart devices and real-time communication. Deployed in voice processing front-ends and edge devices, these algorithms must ensure efficient real-time inference with low computational demands. Traditional edge-based noise suppression often uses MSE-based amplitude spectrum mask training, but this approach has limitations. We introduce a novel projection loss function, diverging from MSE, to enhance noise suppression. This method uses projection techniques to isolate key audio components from noise, significantly improving model performance. For echo cancellation, the function enables direct predictions on LAEC pre-processed outputs, substantially enhancing performance. Our noise suppression model achieves near state-of-the-art results with only 3.1M parameters and 0.4GFlops/s computational load. Moreover, our echo cancellation model outperforms replicated industry-leading models, introducing a new perspective in speech enhancement.