Using Recurrences In Time And Frequency Within U-net Architecture For Speech Enhancement

Abstract

When designing fully-convolutional neural network, there is a trade-off between receptive field size, number of parameters and spatial resolution of features in deeper layers of the network. In this work we present a novel network design based on combination of many convolutional and recurrent layers that solves these dilemmas. We compare our solution with U-nets based models known from the literature and other baseline models on speech enhancement task. We test our solution on TIMIT speech utterances combined with noise segments extracted from NOISEX-92 database and show clear advantage of proposed solution in terms of SDR (signal-to-distortion ratio), SIR (signal-to-interference ratio) and STOI (spectro-temporal objective intelligibility) metrics compared to the current state-of-the-art.

Stats

• citations12
• S2 citations—
• github stars0
• HF likes0
• heat score8.35
• arxiv keygrzywalski2018using

Related papers

• Dilated U-net Based Approach For Multichannel Speech Enhancement From First-order Ambisonics Recordings (2020)0.00
• Towards Speech Enhancement Using A Variational U-net Architecture (2020)7.81
• Improved Speech Enhancement With The Wave-u-net (2018)0.00
• A Fully Recurrent Feature Extraction For Single Channel Speech Enhancement (2020)0.00
• UL-UNAS: Ultra-lightweight U-nets For Real-time Speech Enhancement Via Network Architecture Search (2025)10.26
• Relunet: Relative Channel Fusion U-net For Multichannel Speech Enhancement (2024)0.00
• Single Channel Speech Enhancement Using Temporal Convolutional Recurrent Neural Networks (2020)5.84
• Real-time Streaming Wave-u-net With Temporal Convolutions For Multichannel Speech Enhancement (2021)0.00