EUVIMEDEUVIMED entdecken
Crossref

Nested U-Net-Based Speech Enhancement with Multi-Scale Feature Extraction and Dual-Path Time-Frequency Feature Modeling

Shaik AreefaBegam, Sunnydayal Vanambathina

International Journal of Image, Graphics and Signal Processing · 08.04.2026 · 18 (2) · 107-123

DOI: 10.5815/ijigsp.2026.02.07 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Image, Graphics and Signal Processing, erschienen am 08.04.2026. Verfasst von Shaik AreefaBegam und 1 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Speech enhancement plays a vital role in improving the perceptual quality and intelligibility of speech signals degraded by environmental noise, particularly in modern network-based and signal processing systems. Traditional U-Net architectures capture local spectral details effectively but struggle to model long-range dependencies and may propagate residual noise through skip connections. Transformer-based models provide strong global context modeling but often fail to retain fine-grained spectral cues. To overcome these limitations, this paper presents a Nested U-Net–based network-oriented speech enhancement framework that incorporates Multi-Scale Feature Extraction, Feature Calibration, and a Dual-Path Higher-Order Information Interaction with Time-Frequency Attention module. The Multi-Scale Feature Extraction blocks in both encoder and decoder extract multi-resolution spectral patterns, while the nested topology strengthens hierarchical feature reuse. At the bottleneck, a stack of four Dual-Path Higher-Order Information Interaction with Time-Frequency Attention modules captures long-range temporal and spectral dependencies, and feature calibration adaptively filters encoder features to reduce noise transfer. Extensive experiments on Common Voice and LibriSpeech datasets demonstrate that the proposed model achieves superior perceptual evaluation of speech quality, short-time objective intelligibility, and signal-to-distortion ratio scores, particularly under moderate (0dB) signal-to-noise ratio conditions. The results confirm that the framework provides robust enhancement performance and consistently outperforms several recent state-of-the-art methods in terms of speech quality, intelligibility, and noise suppression.

Zitieren

APA

AreefaBegam, S., & Vanambathina, S. (2026). Nested U-Net-Based Speech Enhancement with Multi-Scale Feature Extraction and Dual-Path Time-Frequency Feature Modeling. International Journal of Image, Graphics and Signal Processing, 18(2), 107-123. https://doi.org/10.5815/ijigsp.2026.02.07

Vancouver

AreefaBegam S, Vanambathina S. Nested U-Net-Based Speech Enhancement with Multi-Scale Feature Extraction and Dual-Path Time-Frequency Feature Modeling. International Journal of Image, Graphics and Signal Processing. 2026;18(2):107-123. doi:10.5815/ijigsp.2026.02.07

BibTeX
@article{AreefaBegam2026,
  author = {AreefaBegam, Shaik and Vanambathina, Sunnydayal},
  title = {Nested U-Net-Based Speech Enhancement with Multi-Scale Feature Extraction and Dual-Path Time-Frequency Feature Modeling},
  journal = {International Journal of Image, Graphics and Signal Processing},
  year = {2026},
  volume = {18},
  number = {2},
  pages = {107--123},
  doi = {10.5815/ijigsp.2026.02.07}
}
Kostenlos starten →