Speech quality and intelligibility are often severely degraded by background noise in communication systems such as hearing aid (HA) and speech recognition technologies, compromising their effective use. In low Signal-to-Noise Ratio (SNR) conditions, various approaches and algorithms are applied to improve speech quality and intelligibility. This study introduces a novel hybrid speech enhancement framework that synergistically integrates Spectral Subtraction (SS) and Discrete Wavelet Transform (DWT) to address limitations of traditional noise reduction techniques. Traditional SS methods generate musical noise artifacts due to static noise estimation, while standard DWT approaches struggle with selective thresholding and static coefficient processing. To overcome these challenges, the proposed SS method incorporates iterative noise estimation, Voice Activity Detection (VAD), minimum statistics for dynamic noise adaptation, Spectral Smoothing and phase-aware spectral reconstruction. Concurrently, in the enhanced DWT method adaptive noise refinement with phase-aware soft thresholding is employed to detail coefficients, and the Spatial and Intensity filter is adapted to the approximation coefficients to improve low-frequency features and retain structural integrity while reducing distortion. The integrated SS-DWT framework significantly improves noise suppression, reduces musical noise artifacts, and enhances signal clarity as it leverages the strengths of both phase-aware spectral reconstruction in improved SS and phase-aware soft thresholding in DWT, particularly in adaptive noise refinement and thresholding. Proposed speech enhancement network evaluated and experimental results show that the hybrid SS-DWT method outperforms existing systems, achieving up to 34.15 dB in SDR, 0.98 in STOI, and 3.84 in PESQ, demonstrating significant improvements in speech quality under various noisy conditions.

At present, pathological voice recognition is mainly based on the classification of pathological voice. However, almost all the researches are based on the single vowel \a\ samples, but few on multi-vowels. In addition, the current researches on multi-vowels recognition are mainly for normal voices, which are unsuitable for the speech recognition of normal and pathological multi-vowels simultaneously. This paper concentrates on developing an accurate and robust feature called enhanced-bark line spectrum pair (E-BLSP) to detect and classify normal and pathological multi-vowels. We explore the impact of E-BLSP feature on recognition performance and propose an effective method based on the combination of three features including E-BLSP for pathological and normal multi-vowels. In this paper, first LSP and difference of adjacent LSP (DAL) features of a vowel are extracted. Then LSP feature is warped at bark domain to get bark line spectrum pair (BLSP). In addition, then E-BLSP feature is calculated by adjusting BLSP using DAL feature. Finally, the adjusted E-BLSP feature and other two traditional features, including linear prediction cepstrum coefficient (LPCC) and mel-frequency cepstrum coefficients (MFCC) are applied to support vector machine (SVM) and deep neural network (DNN) classifiers to explore the classification performance of single feature and feature combinations for pathological and normal vowels /a/, /i/ and /u/. The results show that the highest achieved accuracies for DNN and SVM network are 98.6190% and 96.2693%, while the largest achieved area under curves (AUC) are 0.9925 and 0.9868, correspondingly with the combination of three features including LPCC, MFCC, and E-BLSP.

This study presents a hybrid ensemble learning framework for the joint detection and motor severity prediction of Parkinson’s disease (PD) using biomedical voice features. The proposed architecture integrates a deep multimodal fusion model with dense expert pathways, multi-head self-attention, and multitask output branches to simultaneously perform binary classification and regression. To ensure data quality and improve model generalization, preprocessing steps included outlier removal via Isolation Forest, two-stage feature scaling (RobustScaler followed by MinMaxScaler), and augmentation through polynomial and interaction terms. Borderline-SMOTE was employed to address class imbalance in the classification task. To enhance prediction performance, ensemble learning strategies were applied by stacking outputs from the fusion model with tree-based regressors (Random Forest, Gradient Boosting, and XGBoost), using diverse meta-learners including XGBoost, Ridge Regression, and a deep neural network. Among these, the Stacking Ensemble with XGBoost (SE-XGB) achieved the best results, with an R² of 99.78% and RMSE of 0.3802 for UPDRS regression and 99.37% accuracy for PD classification. Comparative analysis with recent literature highlights the superior performance of our framework, particularly in regression settings. These findings demonstrate the effectiveness of combining advanced feature engineering, deep learning, and ensemble meta-modeling for building accurate and generalizable models in voice-based PD monitoring. This work provides a scalable foundation for future clinical decision support systems.

Deep learning has emerged as a powerful technique for speech enhancement, particularly in security systems where audio signals are often degraded by non-stationary noise. Traditional signal processing methods struggle in such conditions, making it difficult to detect critical sounds like gunshots, alarms, and unauthorized speech. This study investigates a hybrid deep learning framework that combines Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), and Generative Adversarial Networks (GANs) to enhance speech quality and improve sound classification accuracy in noisy security environments. The proposed model is trained and validated using real-world datasets containing diverse noise distortions, including VoxCeleb for benchmarking speech enhancement and UrbanSound8K and ESC-50 for sound classification. Performance is evaluated using industry-standard metrics such as Perceptual Evaluation of Speech Quality (PESQ), Short-Time Objective Intelligibility (STOI), and Signal-to-Noise Ratio (SNR). The architecture includes multi-layered neural networks, residual connections, and dropout regularization to ensure robustness and generalizability. Additionally, the paper addresses key challenges in deploying deep learning models for security applications, such as computational complexity, latency, and vulnerability to adversarial attacks. Experimental results demonstrate that the proposed DNN + GAN-based approach significantly improves speech intelligibility and classification performance in high-interference scenarios, offering a scalable solution for enhancing the reliability of audio-based security systems.