NVH-Denoiser: A Denoising Model Based on CNN+Transformer Structure for NVH Application in New Energy Vehicles Reducer

In addressing the challenges of background noise suppression specifically from the reducers in new energy vehicles (NEVs), we have developed an end-to-end NVH-Denoiser model. This innovative model leverages the strengths of Convolutional Neural Networks (CNNs) for feature extraction and Transformers for capturing temporal dependencies, thus providing a robust framework for noise suppression. Experimental results indicate that the NVH-Denoiser model effectively mitigates background noise emanating from the reducers, demonstrating its capability to meet the stringent NVH requirements in the NEV sector. This advancement promises to significantly enhance the accuracy of noise testing in NEV reducer.