CORTEXA
← Browse
zenodoJournal article2025-07-01

NOISE SUPPRESSION USING DEEP LEARNING (DL) FOR REAL-TIME SPEECH ENHANCEMENT

Khudayberganov Jurabek Davlatboyevich, Berdiyev Alisher Alikulovich

This paper presents a deep learning-based approach to noise suppression in speech signals, comparing convolutional neural networks and recurrent architectures. Inspired by the SEGAN model, a fully convolutional network with residual connections is developed and optimized for TFLite deployment. Results demonstrate effective denoising performance with low latency. Additionally, LSTM models are evaluated, confirming their superiority in handling temporal dependencies in variable-length audio sequences.

View free PDFSource page

Related papers

zenodoJournal article2026-07-27

Hybrid Ensemble Learning for Real-Time Intrusion Detection in Critical Infrastructure Environments.

Nonye Peter Awurum

Abstract: The increasing digitization of critical infrastructure environments has significantly enhanced operational efficiency while simultaneously exposing industrial systems to sophisticated cyber threats. Critical sectors such as oil and gas, energy, transportation, and manuf…

View free PDFSource page
zenodoJournal article2026-07-27

Performance Evaluation of Machine Learning and Deep Learning Models for Real-Time Cyberattack Detection in Oil and Gas Networks

Nonye Peter Awurum

Abstract: The rapid digital transformation of the oil and gas industry has significantly improved operational efficiency through the integration of Information Technology (IT) and Operational Technology (OT) systems. However, this increased connectivity has also expanded the cybe…

View free PDFSource page
zenodoJournal article2026-01-28

A blind binaural real-time model for listening effort evaluated using continuous subjective listening effort rating

Martin Berdau, Daniel-José Alcala Padilla, Thomas Brand, Christian Rollwage, Jan Rennies

This study investigated real-time assessment and modeling of perceived listening effort (LE). The model consists of a binaural front-end, followed by a monaural back-end. As front-end, a novel blind real-time implementation of the binaural speech intelligibility model (BSIM) was…

View free PDFSource page
zenodoJournal article2026-07-27

An AI-Powered Cybersecurity Framework for Real-Time Threat Detection and Resilience in Oil and Gas Critical Infrastructure

Nonye Peter Awurum

Abstract: The rapid digital transformation of the oil and gas industry has accelerated the convergence of Information Technology (IT) and Operational Technology (OT), enabling enhanced operational efficiency, predictive maintenance, and remote asset management. While these techno…

View free PDFSource page
zenodoJournal article2026-07-13

THE IMPACT OF GENERATİVE AI–ENABLED REAL-TİME CONVERSATİONAL MARKETİNG ON DİGİTAL HEALTH EDUCATİON PROCESSES: AN EDUTAİNMENT FRAMEWORK USİNG NO-CODE PLATFORMS

Başal M.

Abstract Advances in digital technologies have fundamentally reshaped health communication and education, giving rise to innovative, interactive, and user-centered learning approaches. In this evolving environment, the edutainment paradigm, which integrates educational content wi…

View free PDFSource page
zenodoJournal article2026-07-28

The Relationships of Principal Leadership, Physical Education Teacher Readiness, and Physical Education Facilities with the Implementation of the Deep Learning Approach

Putri Galuh Anggun Sandra Fatimah, Sugeng Purwanto

This study aimed to examine the relationships between principal leadership, physical education teacher readiness, and physical education facilities and the implementation of the Deep Learning Approach, both individually and collectively. A quantitative correlational survey was co…

View free PDFSource page