Noise-Robust Speech Recognition System based on Multimodal Audio-Visual Approach Using Different Deep Learning Classification Techniques

Application of Deep Learning for Reservoir Porosity Prediction and Self Organizing Map for Lithofacies Prediction

The state-of-energy (SOE) and state-of-health (SOH) are two crucial quotas in the battery management systems, whose accurate estimation is facing challenges by electric vehicles’ (EVs) complexity and changeable external environment. Although the machine learning algorithm can significantly improve the accuracy of battery estimation, it cannot be performed on the vehicle control unit as it requires a large amount of data and computing power. This paper proposes a joint SOE and SOH prediction algorithm, which combines long short-term memory (LSTM), Bi-directional LSTM (Bi-LSTM), and convolutional neural networks (CNNs) for EVs based on vehicle-cloud collaboration. Firstly, the indicator of battery performance degradation is extracted for SOH prediction according to the historical data; the Bayesian optimization approach is applied to the SOH prediction combined with Bi-LSTM. Then, the CNN-LSTM is implemented to provide direct and nonlinear mapping models for SOE. These direct mapping models avoid parameter identification and updating, which are applicable in cases with complex operating conditions. Finally, the SOH correction in SOE estimation achieves the joint estimation with different time scales. With the validation of the National Aeronautics and Space Administration battery data set, as well as the established battery platform, the error of the proposed method is kept within 3%. The proposed vehicle-cloud approach performs high-precision joint estimation of battery SOE and SOH. It can not only use the battery historical data of the cloud platform to predict the SOH but also correct the SOE according to the predicted value of the SOH. The feasibility of vehicle-cloud collaboration is promising in future battery management systems.

This study evaluates the performance of traditional statistical methods, specifically the Seasonal Autoregressive Integrated Moving Average (SARIMA), in comparison to advanced deep learning models—Long Short-Term Memory (LSTM), Bidirectional LSTM (BiLSTM), and Convolutional Neural Network (CNN)—for wind speed forecasting across various regions in Saudi Arabia: Al-Jouf, Abha, Al-Ahsa, and Al-Dawadami. The historical wind speed dataset covering the year 2018 was utilized, with data preprocessing conducted using the Exploratory Data Analysis (EDA) method to ensure consistency and quality. The models were assessed based on three primary error metrics: Mean Absolute Error (MAE), Mean Squared Error (MSE), and Root Mean Square Error (RMSE). Results show that CNN consistently outperforms SARIMA and the other deep learning models, particularly when forecasting stable wind speed patterns. LSTM demonstrates an ability to handle fluctuating wind speeds effectively, while BiLSTM offers advantages in capturing complex bidirectional temporal dependencies. On the other hand, SARIMA generally exhibits lower performance compared to deep learning approaches. The superior performance of CNN is likely attributed to its strength in local feature extraction and handling spatial patterns effectively, which are beneficial for short-term forecasting. These findings provide valuable insights into model selection for wind energy forecasting and can contribute to optimizing renewable energy integration and planning strategies. Future work may explore hybrid model approaches to further enhance forecasting accuracy.

Multimodal multitask deep learning model for Alzheimer’s disease progression detection based on time series data

The prognostics and health management (PHM) plays the main role to handle the risk of failure before its occurrence. Next, it has a broad spectrum of applications including utility networks, energy storage systems (ESS), etc. However, an accurate capacity estimation of batteries in ESS is mandatory for their safe operations and decision making policy. ESS comprises of different storage mechanisms such as batteries, capacitors, etc. Consequently, the measurement of different charging profiles (CPs) has a strong relation to battery capacity. These profiles include temperature (T), voltage (V), and current (I) where the CPs patterns vary as the battery ages with cycles. Consequently, estimating a battery capacity, the conventional methods practice single channel charging profile (SCCP) and hop multiple channel CPs (MCCPs) that cause incorrect battery health estimation. To tackle these issues, this article proposes MCCPs based battery management system (BMS) to estimate batteries health/capacity through the deep learning (DL) concept where the patterns in these CPs are changed as the battery ages with time and cycles. Thus, we deeply investigate both machine learning (ML) and DL based methods to provide a concrete comparative analysis of our method. The adaptive boosting (AB) and support vector regression (SVR) are widely compared with long short-term memory (LSTM), multi-layer perceptron (MLP), bi-directional LSTM (BiLSTM), and convolutional neural network (CNN) to attain the appropriate approach for battery capacity and state of health (SOH) estimation. These approaches have a high learning capability of inter-relation between the battery capacity and variation in CPs patterns. To validate and verify the proposed technique, we use NASA battery dataset and experimentally prove that BiLSTM outperforms all the approaches and obtains the smallest error values for MAE, MSE, RMSE, and MAPE using MCCPs compared to SCCP.

Recent user recognition technologies have focused on biometric signals to remotely identify users in the access management, medical welfare, and public sectors. The electrocardiogram (ECG) signal is an individual’s unique electrophysiological signal generated within the body and is difficult to forge or change; thus, it can be used to uniquely identify the user. Existing user identification systems based on ECG signals detect R peaks according to morphological features and use a fiducial-based segmentation process for data normalization. However, this process does not detect a distinct R peak due to motion artifacts generated by the movement of the subject, thereby degrading identification accuracy. To address the problem of decreasing peak accuracy of the fiducial-based segmentation data, this study proposes a user identification system based on one-dimensional neural networks using periodic non-fiducial-based segmentation data that do not overlap in the time domain. The proposed system comprises a preprocessing step for data denoising, a non-fiducial-based and non-overlap segmentation step, and a step for classifying users using a one-dimensional shallow neural network. In the experiment, when using self-acquired 10-second ECG signal data that were divided into non-fiducial and non-overlapping segments, the long short-term memory (LSTM), bidirectional LSTM, and one-dimensional convolutional neural network (CNN) models achieved accuracies of 92.01%, 93.13%, and 95.51%, respectively, with the 1D CNN model exhibiting the best accuracy. The proposed system demonstrated a 25.48% improvement in accuracy on non-fiducial segmented data compared to the 1D CNN model, which achieved an accuracy of 70.03% on fiducial segmented data. Moreover, when tested on publicly available ECG datasets, including MIT-BIH, ECG-ID, and PTB-XL, the proposed system exhibited a user identification accuracy of over 94%, confirming its superior performance.

Speech production knowledge has been used to enhance the phonetic representation and the performance of automatic speech recognition (ASR) systems successfully. Representations of speech production make simple explanations for many phenomena observed in speech. These phenomena can not be easily analyzed from either acoustic signal or phonetic transcription alone. One of the most important aspects of speech production knowledge is the use of articulatory knowledge, which describes the smooth and continuous movements in the vocal tract. In this paper, we present a new articulatory model to provide available information for rescoring the speech recognition lattice hypothesis. The articulatory model consists of a feature front-end, which computes a voicing feature based on a spectral harmonics correlation (SHC) function, and a back-end based on the combination of deep neural networks (DNNs) and hidden Markov models (HMMs). The voicing features are incorporated with standard Mel frequency cepstral coefficients (MFCCs) using heteroscedastic linear discriminant analysis (HLDA) to compensate the speech recognition accuracy rates. Moreover, the advantages of two different models are taken into account by the algorithm, which retains deep learning properties of DNNs, while modeling the articulatory context powerfully through HMMs. Mandarin speech recognition experiments show the proposed method achieves significant improvements in speech recognition performance over the system using MFCCs alone.

This paper proposes a multi-layered anomaly detection scheme to train feature extraction and to test anomaly prediction by using Convolutional Neural Networks (CNNs) layer, Bidirectional and Unidirectional Long Short-Term Memory (LSTM) Recurrent Neural Networks (RNNs), which is one of a novel deep architecture named stacked convolutional bidirectional LSTM network (SCB-LSTM). In the proposed model, the stacked CNNs perform feature extraction of vibration sensor signal patterns, and the result is used to feature learning with the stacked bidirectional LSTMs (SB-LSTMs). After this procedure, the stacked unidirectional LSTMs (SU-LSTMs) enhance the feature learning, and a regression layer finally predicts anomaly detections. The experimental results of bearing data not only show the accuracy of the proposed model in anomaly detection for rotating machinery diagnostics, but also suggest the better performance than other state-of-the-art algorithms such as a plain uni-LSTM or Bi-LSTM.

Parts of Speech (pos) tagging corresponds to techniques of tagging accurate pos tags of each word of a sentence. It is a pivotal and indispensable part of sentiment analysis, synthesizing text to speech, analyzing textual data and many more Natural Language Processing (NLP) tasks. Though there are many researches done in the field of languages like English, Chinese etc., it is still comparatively unexplored for Bangla. An excellent pos tagger in Bangla language can accelerate NLP of Bangla textual data. In this study, we have developed a novel hybrid Deep Learning (DL) model with two parallel input networks with Convolutional Neural Network (CNN), and Bidirectional Long Short Term Memory (BiLSTM), and an output network of CNN. CNN works well regarding the local dependencies of words in sentences, and BiLSTM works well with the global dependencies of words regarding the context in a sentence. No other work has been done with local features of textual data along with long term word dependencies in Bangla pos tagging studies. The model can work with any external Word Embeddings. In this study, it is implemented with both Embedding Layer and Word2Vec layer. In the evaluation process, Mathews Correlation Coefficient (MCC), Precision-Recall curve are used along with accuracy and F1-score due to the dataset's imbalanced characteristic. Our proposed model accompanied by Word2Vec layer has outperformed Vanilla Recurrent Neural Network (RNN), Gated Recurrent Units (GRU), Long Short Term Memory (LSTM), BiLSTM networks, and also one recent study using the same dataset with the highest accuracy of 0.974, F1-score of 0.883, and MCC value of 0.936.

The utilization of Convolutional Neural Networks (CNNs) and Bidirectional Long Short-Term Memory (BLSTM) networks in image captioning has significantly enhanced the quality and relevance of generated captions. In this approach, a CNN serves as the encoder to extract meaningful features from the input image, capturing its visual information. These features are then fed into a BLSTM network, acting as the decoder, which processes the features in a bidirectional manner to generate descriptive and coherent captions by considering both past and future context. The model is trained on a dataset of images and corresponding captions, fine-tuning the parameters to optimize the accuracy of caption generation. This combined approach effectively leverages the strengths of CNNs and BLSTMs, resulting in more detailed and contextually appropriate descriptions for the given images.

Predicting remaining useful life (RUL) is crucial for tool condition monitoring (TCM) systems. Inaccurate predictions can lead to premature tool replacements or excessive usage, resulting in resource wastage and potential equipment failures. This study introduces a novel tool RUL prediction method that integrates the enhanced northern goshawk optimization (MSANGO) algorithm with a bidirectional long short-term memory (BiLSTM) network. Initially, key statistical features are extracted from collected signal data using multivariate variational mode decomposition. This is followed by effective feature reduction, facilitated by the uniform information coefficient and Mann–Kendall trend tests. The RUL predictions are subsequently refined through a BiLSTM network, with the MSANGO algorithm optimizing the network parameters. Comparative evaluations with BiLSTM, BiGRU, and NGO-BiLSTM models, as well as tests on real-world datasets, demonstrate this method’s superior accuracy and generalizability in RUL prediction, enhancing the efficacy of tool management systems.

Investigation of acoustic and visual features for pig cough classification

Speech recognition can be defined as the process of converting voice signals into the ranks of the word, by applying a specific algorithm that is implemented in a computer program. The research of speech recognition in Indonesia is relatively limited. This paper has studied methods of feature extraction which is the best among the Linear Predictive Coding (LPC) and Mel Frequency Cepstral Coefficients (MFCC) for speech recognition in Indonesian language. This is important because the method can produce a high accuracy for a particular language does not necessarily produce the same accuracy for other languages, considering every language has different characteristics. Thus this research hopefully can help further accelerate the use of automatic speech recognition for Indonesian language. There are two main processes in speech recognition, feature extraction and recognition. The method used for comparison feature extraction in this study is the LPC and MFCC, while the method of recognition using Hidden Markov Model (HMM). The test results showed that the LPC method is better than MFCC in Indonesian language speech recognition.

Prediction of PM2.5 concentration in urban agglomeration of China by hybrid network model

This paper proposes a hybrid modeling framework that synergistically integrates LSTM (Long Short-Term Memory) networks with LightGBM and CatBoost for stock price prediction. We systematically preprocess time-series financial data and conduct comprehensive evaluations using seven classical models: Artificial Neural Networks (ANNs), Convolutional Neural Networks (CNNs), Bidirectional LSTM (BiLSTM), vanilla LSTM, XGBoost, LightGBM, and standard Neural Networks (NNs). Through rigorous comparison of performance metrics, including MAE, R2, MSE, and RMSE, we establish baseline references across multiple temporal scales. Building on these empirical benchmarks, we introduce a novel ensemble architecture that harmonizes the complementary strengths of sequential and tree-based models. Experimental results demonstrate that our integrated framework achieves 10-15% improvement in predictive accuracy compared to individual constituent models, particularly in reducing error volatility during market regime transitions. These findings validate the underexplored potential of heterogeneous ensemble strategies in financial forecasting and establish methodological foundations for developing adaptive prediction systems in non-stationary market environments. The proposed architecture's modular design enables seamless integration of emerging machine learning components, suggesting promising directions for future research in computational finance.