Search papers, labs, and topics across Lattice.
This paper introduces SEMamba++, a speech restoration framework that enhances the SEMamba architecture by incorporating speech-specific inductive biases. The core innovation is the Frequency GLP block for efficient frequency feature extraction and a multi-resolution parallel time-frequency dual-processing block to capture diverse spectral patterns. Experiments demonstrate that SEMamba++ achieves state-of-the-art speech restoration performance with computational efficiency.
SEMamba++ significantly improves speech restoration by cleverly integrating frequency-domain inductive biases into a state-space model, outperforming existing methods while maintaining efficiency.
General speech restoration demands techniques that can interpret complex speech structures under various distortions. While State-Space Models like SEMamba have advanced the state-of-the-art in speech denoising, they are not inherently optimized for critical speech characteristics, such as spectral periodicity or multi-resolution frequency analysis. In this work, we introduce an architecture tailored to incorporate speech-specific features as inductive biases. In particular, we propose Frequency GLP, a frequency feature extraction block that effectively and efficiently leverages the properties of frequency bins. Then, we design a multi-resolution parallel time-frequency dual-processing block to capture diverse spectral patterns, and a learnable mapping to further enhance model performance. With all our ideas combined, the proposed SEMamba++ achieves the best performance among multiple baseline models while remaining computationally efficient.
