Transformer noise removal method based on IMGO-CEEMD

This paper conducts an experimental study on the noise filtering of transformer voiceprint signals. The results show that the proposed method has increased the signal-to-noise ratio of the noise by 2.2 to 3.5 decibels, and reduced the root mean square error by 8% to 25%. Therefore, this method can consistently improve the denoising effect in complex noise environments and provide an effective foundation for high-precision voiceprint recognition.