DualStream Contextual Fusion Network: Efficient Target Speaker Extraction by Leveraging Mixture and Enrollment Interactions

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Xue, Ke, Fan, Rongfei, Yu, Shanping, Sun, Chang, An, Jianping
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866929711853076480
author Xue, Ke
Fan, Rongfei
Yu, Shanping
Sun, Chang
An, Jianping
author_facet Xue, Ke
Fan, Rongfei
Yu, Shanping
Sun, Chang
An, Jianping
contents Target speaker extraction focuses on extracting a target speech signal from an environment with multiple speakers by leveraging an enrollment. Existing methods predominantly rely on speaker embeddings obtained from the enrollment, potentially disregarding the contextual information and the internal interactions between the mixture and enrollment. In this paper, we propose a novel DualStream Contextual Fusion Network (DCF-Net) in the time-frequency (T-F) domain. Specifically, DualStream Fusion Block (DSFB) is introduced to obtain contextual information and capture the interactions between contextualized enrollment and mixture representation across both spatial and channel dimensions, and then rich and consistent representations are utilized to guide the extraction network for better extraction. Experimental results demonstrate that DCF-Net outperforms state-of-the-art (SOTA) methods, achieving a scale-invariant signal-to-distortion ratio improvement (SI-SDRi) of 21.6 dB on the benchmark dataset, and exhibits its robustness and effectiveness in both noise and reverberation scenarios. In addition, the wrong extraction results of our model, called target confusion problem, reduce to 0.4%, which highlights the potential of DCF-Net for practical applications.
format Preprint
id arxiv_https___arxiv_org_abs_2502_08191
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle DualStream Contextual Fusion Network: Efficient Target Speaker Extraction by Leveraging Mixture and Enrollment Interactions
Xue, Ke
Fan, Rongfei
Yu, Shanping
Sun, Chang
An, Jianping
Sound
Audio and Speech Processing
Target speaker extraction focuses on extracting a target speech signal from an environment with multiple speakers by leveraging an enrollment. Existing methods predominantly rely on speaker embeddings obtained from the enrollment, potentially disregarding the contextual information and the internal interactions between the mixture and enrollment. In this paper, we propose a novel DualStream Contextual Fusion Network (DCF-Net) in the time-frequency (T-F) domain. Specifically, DualStream Fusion Block (DSFB) is introduced to obtain contextual information and capture the interactions between contextualized enrollment and mixture representation across both spatial and channel dimensions, and then rich and consistent representations are utilized to guide the extraction network for better extraction. Experimental results demonstrate that DCF-Net outperforms state-of-the-art (SOTA) methods, achieving a scale-invariant signal-to-distortion ratio improvement (SI-SDRi) of 21.6 dB on the benchmark dataset, and exhibits its robustness and effectiveness in both noise and reverberation scenarios. In addition, the wrong extraction results of our model, called target confusion problem, reduce to 0.4%, which highlights the potential of DCF-Net for practical applications.
title DualStream Contextual Fusion Network: Efficient Target Speaker Extraction by Leveraging Mixture and Enrollment Interactions
topic Sound
Audio and Speech Processing
url https://arxiv.org/abs/2502.08191