LSTMSE-Net: Long Short Term Speech Enhancement Network for Audio-visual Speech Enhancement
This work addresses speech enhancement for noisy environments by combining audio and visual data, but it appears incremental as it builds on existing multi-modal approaches with specific improvements.
The paper tackles audio-visual speech enhancement by proposing LSTMSE-Net, which integrates visual and audio features to improve speech quality, achieving gains of 0.06 in SISDR, 0.03 in STOI, and 1.32 in PESQ over a baseline.
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}.