LSTMSE-Net: Long Short Term Speech Enhancement Network for Audio-visual Speech Enhancement

TL;DR

LSTMSE-Net is a novel audio-visual speech enhancement network that combines visual and audio features to significantly improve speech quality and intelligibility in noisy environments.

Contribution

The paper introduces LSTMSE-Net, a new multi-modal speech enhancement architecture that leverages visual and audio data with interpolation techniques for improved performance.

Findings

Outperforms baseline in SISDR, STOI, and PESQ metrics.

Effectively utilizes visual features for speech enhancement.

Achieves state-of-the-art results in AVSE challenge.

Abstract

In this paper, we propose long short term memory speech enhancement network (LSTMSE-Net), an audio-visual speech enhancement (AVSE) method. This innovative method leverages the complementary nature of visual and audio information to boost the quality of speech signals. Visual features are extracted with VisualFeatNet (VFN), and audio features are processed through an encoder and decoder. The system scales and concatenates visual and audio features, then processes them through a separator network for optimized speech enhancement. The architecture highlights advancements in leveraging multi-modal data and interpolation techniques for robust AVSE challenge systems. The performance of LSTMSE-Net surpasses that of the baseline model from the COG-MHEAR AVSE Challenge 2024 by a margin of 0.06 in scale-invariant signal-to-distortion ratio (SISDR), $0.03$ in short-time objective intelligibility (STOI), and $1.32$ in perceptual evaluation of speech quality (PESQ). The source code of the proposed LSTMSE-Net is available at \url{https://github.com/mtanveer1/AVSEC-3-Challenge}.