EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

A Comparative Analysis of Deep-Learning-Based Speech Enhancement Models: Assessing Biometric Speaker Verification in Real-World Noisy Environments

Md Jahangir Alam Khondkar, Ajan Ahmed, Stephanie Schuckers, Masudul H. Imtiaz

Big Data and Cognitive Computing · 23.03.2026 · 10 (3) · 98

DOI: 10.3390/bdcc10030098 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Big Data and Cognitive Computing, online erschienen am 23.03.2026. Verfasst von Md Jahangir Alam Khondkar und 3 weiteren Autor:innen. Die Arbeit nennt 34 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Center for Identification Technology Research, National Science Foundation.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Speech enhancement through denoising is essential for maintaining signal intelligibility and quality in biometric speaker verification pipelines that operate in acoustically adverse conditions. Despite the proliferation of deep learning (DL) architectures for speech denoising, simultaneously optimizing noise attenuation, perceptual fidelity, and speaker-identity preservation remains an open problem. We address this gap by benchmarking three architecturally distinct DL-based enhancement models—Wave-U-Net, CMGAN, and U-Net—on three independent, domain-diverse corpora (SpEAR, VPQAD, and Clarkson) that the models never encountered during training and by introducing commercial-grade VeriSpeak speaker-verification scores as a biometric evaluation dimension absent from prior comparative studies. Our experiments reveal a clear three-way trade-off: U-Net achieves the highest signal-to-noise ratio (SNR) gains (+61.44% on SpEAR, +67.05% on VPQAD, +235.3% on Clarkson) but sacrifices naturalness; CMGAN yields the best perceptual evaluation of speech quality (PESQ) values (3.33, 1.35, and 2.50, respectively), favoring listening-comfort applications; and Wave-U-Net delivers the strongest biometric fidelity (VeriSpeak improvements of +11.63%, +30.22%, and +29.24%) while offering competitive perceptual quality. These results highlight that model selection must be driven by the target deployment scenario and provide actionable guidance for improving biometric verification robustness under real-world noise.

Förderung

  • Center for Identification Technology Research
  • National Science Foundation

Zitieren

APA

Khondkar, M. J. A., Ahmed, A., Schuckers, S., & Imtiaz, M. H. (2026). A Comparative Analysis of Deep-Learning-Based Speech Enhancement Models: Assessing Biometric Speaker Verification in Real-World Noisy Environments. Big Data and Cognitive Computing, 10(3), 98. https://doi.org/10.3390/bdcc10030098

Vancouver

Khondkar MJA, Ahmed A, Schuckers S, Imtiaz MH. A Comparative Analysis of Deep-Learning-Based Speech Enhancement Models: Assessing Biometric Speaker Verification in Real-World Noisy Environments. Big Data and Cognitive Computing. 2026;10(3):98. doi:10.3390/bdcc10030098

BibTeX
@article{Khondkar2026,
  author = {Khondkar, Md Jahangir Alam and Ahmed, Ajan and Schuckers, Stephanie and Imtiaz, Masudul H.},
  title = {A Comparative Analysis of Deep-Learning-Based Speech Enhancement Models: Assessing Biometric Speaker Verification in Real-World Noisy Environments},
  journal = {Big Data and Cognitive Computing},
  year = {2026},
  volume = {10},
  number = {3},
  pages = {98},
  doi = {10.3390/bdcc10030098}
}
Kostenlos starten →