Deep Active Speech Cancellation with Mamba-Masking Network

ICLR 2026 Conference Submission18143 Authors

Keywords: Active Noise Cancellation, Mamba

Abstract: We present a novel deep learning network for Active Speech Cancellation (ASC), advancing beyond Active Noise Cancellation (ANC) methods by effectively canceling both noise and speech signals. The proposed Mamba-Masking architecture introduces a masking mechanism that directly interacts with the encoded reference signal, enabling adaptive and precisely aligned anti-signal generation—even under rapidly changing, high-frequency conditions, as commonly found in speech. Complementing this, a multi-band segmentation strategy further improves phase alignment across frequency bands. Additionally, we introduce an optimization-driven loss function that provides near-optimal supervisory signals for anti-signal generation. Experimental results demonstrate substantial performance gains, achieving up to 7.2dB improvement in ANC scenarios and 6.2dB in ASC, significantly outperforming existing methods. For reproducibility, our code along with audio samples can be found on the supplementary materials.

Primary Area: applications to computer vision, audio, language, and other modalities

Submission Number: 18143