Saved in:
Bibliographic Details
Main Authors: Li, Kai, Gao, Kejun, Hu, Xiaolin
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2509.23610
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915852842958848
author Li, Kai
Gao, Kejun
Hu, Xiaolin
author_facet Li, Kai
Gao, Kejun
Hu, Xiaolin
contents Audio-visual speech separation (AVSS) methods leverage visual cues to extract target speech and have demonstrated strong separation quality in noisy acoustic environments. However, these methods usually involve a large number of parameters and require high computational cost, which is unacceptable in many applications where speech separation serves as only a preprocessing step for further speech processing. To address this issue, we propose an efficient AVSS method, named Dolphin. For visual feature extraction, we develop DP-LipCoder, a dual-path lightweight video encoder that transforms lip-motion into discrete audio-aligned semantic tokens. For audio separation, we construct a lightweight encoder-decoder separator, in which each layer incorporates a global-local attention (GLA) block to efficiently capture multi-scale dependencies. Experiments on three benchmark datasets showed that Dolphin not only surpassed the current state-of-the-art (SOTA) model in separation quality but also achieved remarkable improvements in efficiency: over 50% fewer parameters, more than 2.4x reduction in MACs, and over 6x faster GPU inference speed. These results indicate that Dolphin offers a practical and deployable solution for high-performance AVSS in real-world scenarios. Our code and demo page are publicly available at http://cslikai.cn/Dolphin/.
format Preprint
id arxiv_https___arxiv_org_abs_2509_23610
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Efficient Audio-Visual Speech Separation with Discrete Lip Semantics and Multi-Scale Global-Local Attention
Li, Kai
Gao, Kejun
Hu, Xiaolin
Sound
Computer Vision and Pattern Recognition
Audio-visual speech separation (AVSS) methods leverage visual cues to extract target speech and have demonstrated strong separation quality in noisy acoustic environments. However, these methods usually involve a large number of parameters and require high computational cost, which is unacceptable in many applications where speech separation serves as only a preprocessing step for further speech processing. To address this issue, we propose an efficient AVSS method, named Dolphin. For visual feature extraction, we develop DP-LipCoder, a dual-path lightweight video encoder that transforms lip-motion into discrete audio-aligned semantic tokens. For audio separation, we construct a lightweight encoder-decoder separator, in which each layer incorporates a global-local attention (GLA) block to efficiently capture multi-scale dependencies. Experiments on three benchmark datasets showed that Dolphin not only surpassed the current state-of-the-art (SOTA) model in separation quality but also achieved remarkable improvements in efficiency: over 50% fewer parameters, more than 2.4x reduction in MACs, and over 6x faster GPU inference speed. These results indicate that Dolphin offers a practical and deployable solution for high-performance AVSS in real-world scenarios. Our code and demo page are publicly available at http://cslikai.cn/Dolphin/.
title Efficient Audio-Visual Speech Separation with Discrete Lip Semantics and Multi-Scale Global-Local Attention
topic Sound
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2509.23610