PLDNet: PLD-Guided Lightweight Deep Network Boosted by Efficient Attention for Handheld Dual-Microphone Speech Enhancement

TL;DR

PLDNet is a lightweight dual-microphone speech enhancement model that combines power level difference guidance with an attention-augmented U-Net, achieving high performance with significantly reduced computational cost for mobile devices.

Contribution

This paper introduces PLDNet, a novel lightweight deep network that integrates PLD guidance and a new attention module for efficient speech enhancement on mobile phones.

Findings

Achieves competitive speech enhancement performance.

Reduces computational cost by over 90%.

Suitable for low-power mobile devices.

Abstract

Low-complexity speech enhancement on mobile phones is crucial in the era of 5G. Thus, focusing on handheld mobile phone communication scenario, based on power level difference (PLD) algorithm and lightweight U-Net, we propose PLD-guided lightweight deep network (PLDNet), an extremely lightweight dual-microphone speech enhancement method that integrates the guidance of signal processing algorithm and lightweight attention-augmented U-Net. For the guidance information, we employ PLD algorithm to pre-process dual-microphone spectrum, and feed the output into subsequent deep neural network, which utilizes a lightweight U-Net with our proposed gated convolution augmented frequency attention (GCAFA) module to extract desired clean speech. Experimental results demonstrate that our proposed method achieves competitive performance with recent top-performing models while reducing computational cost by over 90%, highlighting the potential for low-complexity speech enhancement on mobile phones.