PrimeK-Net: Multi-scale Spectral Learning via Group Prime-Kernel Convolutional Neural Networks for Single Channel Speech Enhancement

TL;DR

PrimeK-Net introduces a multi-scale spectral learning approach using group prime-kernel convolutions, achieving state-of-the-art speech enhancement performance with high efficiency and fewer parameters.

Contribution

The paper presents PrimeK-Net, a novel CNN-based framework with group prime-kernel convolutions and efficient modules for improved speech enhancement.

Findings

Achieves SOTA PESQ score of 3.61 on VoiceBank+Demand dataset.

Uses only 1.41 million parameters, demonstrating efficiency.

Outperforms existing methods in spectral feature extraction and computational cost.

Abstract

Single-channel speech enhancement is a challenging ill-posed problem focused on estimating clean speech from degraded signals. Existing studies have demonstrated the competitive performance of combining convolutional neural networks (CNNs) with Transformers in speech enhancement tasks. However, existing frameworks have not sufficiently addressed computational efficiency and have overlooked the natural multi-scale distribution of the spectrum. Additionally, the potential of CNNs in speech enhancement has yet to be fully realized. To address these issues, this study proposes a Deep Separable Dilated Dense Block (DSDDB) and a Group Prime Kernel Feedforward Channel Attention (GPFCA) module. Specifically, the DSDDB introduces higher parameter and computational efficiency to the Encoder/Decoder of existing frameworks. The GPFCA module replaces the position of the Conformer, extracting deep temporal and frequency features of the spectrum with linear complexity. The GPFCA leverages the proposed Group Prime Kernel Feedforward Network (GPFN) to integrate multi-granularity long-range, medium-range, and short-range receptive fields, while utilizing the properties of prime numbers to avoid periodic overlap effects. Experimental results demonstrate that PrimeK-Net, proposed in this study, achieves state-of-the-art (SOTA) performance on the VoiceBank+Demand dataset, reaching a PESQ score of 3.61 with only 1.41M parameters.