FN-SSL: Full-Band and Narrow-Band Fusion for Sound Source Localization
Audio and Speech Processing
2023-06-01 v1
Abstract
Extracting direct-path spatial features is critical for sound source localization in adverse acoustic environments. This paper proposes a full-band and narrow-band fusion network for estimating direct-path inter-channel phase difference (DP-IPD) from microphone signals. The alternating full-band and narrow-band layers are responsible for learning the full-band correlation and narrow-band extraction of DP-IPD, respectively. Experiments show that the proposed network noticeably outperforms other advanced methods on both simulated and real-world data.
Keywords
Cite
@article{arxiv.2305.19610,
title = {FN-SSL: Full-Band and Narrow-Band Fusion for Sound Source Localization},
author = {Yabo Wang and Bing Yang and Xiaofei Li},
journal= {arXiv preprint arXiv:2305.19610},
year = {2023}
}