A Multi-dimensional Deep Structured State Space Approach to Speech Enhancement Using Small-footprint Models

Abstract

We propose a multi-dimensional structured state space (S4) approach to speech enhancement. To better capture the spectral dependencies across the frequency axis, we focus on modifying the multi-dimensional S4 layer with whitening transformation to build new small-footprint models that also achieve good performance. We explore several S4-based deep architectures in time (T) and time-frequency (TF) domains. The 2-D S4 layer can be considered a particular convolutional layer with an infinite receptive field although it utilizes fewer parameters than a conventional convolutional layer. Evaluated on the VoiceBank-DEMAND data set, when compared with the conventional U-net model based on convolutional layers, the proposed TF-domain S4-based model is 78.6% smaller in size, yet it still achieves competitive results with a PESQ score of 3.15 with data augmentation. By increasing the model size, we can even reach a PESQ score of 3.18.

arXiv commentAccepted to Interspeech 2023. Code will be released at https://github.com/Kuray107/S4ND-U-Net_speech_enhancement

References in corpus (8)

• On the difficulty of training Recurrent Neural Networks
• Conv-TasNet: Surpassing Ideal Time-Frequency Magnitude Masking for Speech Separation
• Efficiently Modeling Long Sequences with Structured State Spaces
• Optimal whitening and decorrelation
• Speech Enhancement and Dereverberation with Diffusion-based Generative Models
• Variational Autoencoder for Speech Enhancement with a Noise-Aware Encoder
• It's Raw! Audio Generation with State-Space Models
• S-DCCRN: Super Wide Band DCCRN with learnable complex feature for speech enhancement