SICRN: Advancing Speech Enhancement Through State Space Model And Inplace Convolution Techniques

Abstract

Speech enhancement aims to improve speech quality and intelligibility, especially in noisy environments where background noise degrades speech signals. Currently, deep learning methods achieve great success in speech enhancement, e.g. the representative convolutional recurrent neural network (CRN) and its variants. However, CRN typically employs consecutive downsampling and upsampling convolution for frequency modeling, which destroys the inherent structure of the signal over frequency. Additionally, convolutional layers lacks of temporal modelling abilities. To address these issues, we propose an innovative module combing a State space model and Inplace Convolution (SIC), and to replace the conventional convolution in CRN, called SICRN. Specifically, a dual-path multidimensional State space model captures the global frequencies dependency and long-term temporal dependencies. Meanwhile, the 2D-inplace convolution is used to capture the local structure, which abandons the downsampling a

Stats

• citations10
• S2 citations—
• github stars0
• HF likes0
• heat score7.81
• arxiv keyzhao2024sicrn

Related papers

• DCCRN: Deep Complex Convolution Recurrent Network For Phase-aware Speech Enhancement (2020)20.78
• Inplace Gated Convolutional Recurrent Neural Network For Dual-channel Speech Enhancement (2021)0.00
• Complex Spectral Mapping With Attention Based Convolution Recurrent Neural Network For Speech Enhancement (2021)0.00
• Single Channel Speech Enhancement Using Temporal Convolutional Recurrent Neural Networks (2020)5.84
• What Do Neural Networks Listen To? Exploring The Crucial Bands In Speech Enhancement Using Sinc-convolution (2024)2.26
• Wavecrn: An Efficient Convolutional Recurrent Neural Network For End-to-end Speech Enhancement (2020)14.02
• A Multi-dimensional Deep Structured State Space Approach To Speech Enhancement Using Small-footprint Models (2023)9.23
• Real-time Monaural Speech Enhancement With Short-time Discrete Cosine Transform (2021)0.00