Embedded Implementation of Real-Time Voice Command Recognition on PIC Microcontroller

Deep features generated from deep learning models contain more information for music classification than short-term features. This paper uses a long-short term memory (LSTM) model to generate deep features and achieve music genre classification. Firstly, two short-term features of Zero crossing rate (ZCR) and mel-frequency spectral coefficients (MFCC) are extracted from music in digital form, which is a time-domain feature and frequency-domain feature, respectively. Then these two features are fed to LSTM to generate deep features. Finally, we use support vector machine (SVM) and k-nearest neighbors (KNN) respectively to classify the music genre based on these deep features. Experimental results show that using LSTM can significantly increase the accuracy of music genre classification.

Cloud computing and big data: Technologies and applications

In this work, we have presented a novel simple scheme for classifying audio speech signals into male speech and female speech. In the context of content-based multimedia indexing gender identification based on speech signal is an important task. Some popular salient low level time-domain acoustic features which are very closely related to the physical properties of source audio signal like zero crossing rate (ZCR), short time energy (STE) along with spectral flux, a low level frequency domain feature, are used for this discrimination. RANSAC and Neural-Net has been used as classifier. The experimental result exhibits the efficiency of the proposed scheme.

Epileptic seizure is a common neurological syndrome wherein patients suffer from repeating seizures. It happens due to many provoking reasons namely genetic, physiologic, brain damage, etc. In this paper, time domain and frequency domain feature set are derived from EEG data and classified using Convolutional Neural Network (CNN) for detecting epileptic seizures. The benchmark EEG dataset Bonn is experimented to compare and examine the result of forthcoming methods. The proposed methodologies have been evaluated using the metrics such as sensitivity, specificity, classification accuracy and execution using two different benchmark datasets and results are compared with exiting methods. It is learned that among various methods the proposed approach using convolutional neural network with time domain features provides enhanced performance measured result. As the time domain features are computationally inexpensive to extract and also required reduced storage, so it can be used in real time automatic seizure detection and monitoring using wireless systems.

Speech and Music Classification and Separation: A Review

The focus of this paper is on classification of different vehicles using sound emanated from the vehicles. In this paper,quadratic discriminant analysis classifies audio signals of passing vehicles to bus, car, motor, and truck categories based on features such as short time energy, average zero cross rate, and pitch frequency of periodic segments of signals. Simulation results show that just by considering high energy feature vectors, better classification accuracy can be achieved due to the correspondence of low energy regions with noises of the background. To separate these elements, short time energy and average zero cross rate are used simultaneously.In our method,we have used a few features which are easy to be calculated in time domain and enable practical implementation of efficient classifier. Although, the computation complexity is low, the classification accuracy is comparable with other classification methods based on long feature vectors reported in literature for this problem.

Introduction: In order to assess how well conventional and hybrid pitch detection techniques perform in speech processing applications, a comparative analysis of the two types of methods is conducted. Method: A proposed hybrid approach, Proposed PEF+CEP, is examined alongside five traditional algorithms, namely Normalized Correlation Function (NCF), Pitch Estimation Filter (PEF), Log- Harmonic Summation (LHS), Summation of Residual Harmonics (SRH) and Cepstrum Pitch Determination (CEP). The effectiveness is evaluated using performance metrics like accuracy, specificity, sensitivity, and Gross Pitch Error (GPE). Results and Discussion: Our findings show that the accuracy and specificity of the traditional methods are impressive; the accuracy and sensitivity of the suggested hybrid method surpass their performance, with an astounding 98.8% accuracy and 99.2% sensitivity. Conclusion: Furthermore, the Proposed PEF+CEP method is a promising solution for accurate and dependable pitch detection in speech processing applications because it strikes a strong balance between computational efficiency and robustness. These results open up new avenues for research in the field of speech processing and demonstrate the potential of hybrid approaches.

Turmeric and curcuma are herbs that are often used in medicine and cooking. However, their similar shapes and colours make it difficult for people, especially in Southwest Papua, to distinguish between them directly. According to the Central Statistics Agency (BPS) in 2023, turmeric production reached 18,302 units, far higher than turmeric, which only reached 2,950 units. Based on field interviews in Southwest Papua, more than 60% of respondents had difficulty distinguishing turmeric from turmeric. To address this issue, this research develops an Android-based classification system by integrating the Fuzzy Tsukamoto algorithm with Convolutional Neural Network (CNN) models. Five CNN models VGG16, MobileNetV2, NASNetMobile, EfficientNetB2, and EfficientNetB3 were selected based on their balance between computational efficiency (MobileNetV2, NASNetMobile), depth and proven stability (VGG16), and modern scalable architectures (EfficientNetB2 and B3). Each model was combined with fuzzy logic to enhance classification accuracy. he dataset consisted of 800 images of curcuma and turmeric obtained from Kaggle and field collections. The data were divided into training, validation, and testing sets, and augmented through a series of transformations including rescaling to a range of 0 to 1, rotation up to 40 degrees, horizontal shift of 20%, angular distortion (shear) of 20%, zoom up to 30%, horizontal flipping, and brightness adjustment. Empty areas generated during augmentation were filled using the nearest pixel value with the ‘nearest’ mode to preserve image integrity. Training was performed using the AdamW optimizer and fine-tuning. Model evaluation employed accuracy, precision, recall, F1-score, and confusion matrix metrics. The results showed that the VGG16 model performed best, achieving 97% accuracy, 98% precision, 97% recall, and 98% F1-score, as confirmed by the classification report and confusion matrix. This model was also the most stable when tested on the Android system, while EfficientNetB2 and B3 produced less satisfactory outcomes. These findings demonstrate that combining CNN and Fuzzy Tsukamoto improves the classification accuracy of images with high visual similarity. The proposed system has the potential to be applied as a direct plant identification tool in the field and can be further extended to classify other visually similar plants

People with disabilities suffer from inability to communicate with their surroundings, so Human-Computer Interaction (HCI) technologies are used to have a means of communication for people with disabilities with their surroundings. HCI is an emerging technology in the disciplines of Artificial Intelligence and Biomedical Engineering. To power an external device, HCI technology uses several basic signals such as ECG, EMG, and EEG. Electrooculography (EOG) is a technique for measuring the potential difference between the cornea and the retina located between the front and back of the human eye, and the main application of EOG is to determine the directions of different eye movements. This study aims to assess eye movement for communication by persons with disabilities using electrocardiogram (EOG) data. In this study, the Supporting Vector Machine (SVM) and Long- Short term memory (LSTM) classification techniques was used and two types of features (statistical and time domain features) were used. Classification accuracy was 90.7% and 93.9% when using SVM with statistical domain and time domain features, respectively ,whereas Classification accuracy was 90.1% when using LSTM .

Passive acoustic monitoring (PAM) is generally usedto extract acoustic signals produced by cetaceans. However, the large data volume from the PAM process is better analyzed using an automated technique such as the hidden Markovmodels (HMM). In this paper, the HMM is used as a detection and classification technique due to its robustness and low time complexity. Nonetheless, certain parameters, such as the choice of features to be extracted from the signal, the frame duration, and the number of states affect the performance of the model. Theresults show that HMM exhibits best performances as the number of states increases with short frame duration. However, increasing the number of states creates more computational complexity in the model. The inshore Bryde's whales produce short pulse calls with distinct signal features, which are observable in the time-domain. Hence, a time-domain feature vector is utilized to reduce the complexity of the HMM. Simulation results also show that average power as a time-domain feature vector provides the best performance compared to other feature vectors for detecting the short pulse call of inshore Bryde's whales based on the HMM technique. More so, the extracted features such as the average power, mean, and zero-crossing rate, are combined to form a single 3-dimensional vector (PaMZ). The PaMZ-HMM shows improved performance and reduced complexity over existing feature extraction techniques such as Mel-scale frequency cepstral coefficients (MFCC) and linear predictive coding (LPC). Thus, making the PaMZ-HMM suitable for real-time detection.

Remote sensing data have long been the primary source for land cover map derivation. Nevertheless, for countries within haze-affected regions such as Malaysia, the existence of haze in the atmosphere tends to degrade the data quality. Such scenario is due to attenuation of recorded reflectances in which consequently affects the land cover classification task prior to the map derivation. This study aims to determine the effects of haze on the accuracy of land cover classification. Landsat-5 TM (Thematic Mapper) satellite data over the district of Klang, located in the state of Selangor, Malaysia were used. To account for haze effects, the study made use the Landsat datasets that have been integrated with haze layers. Maximum Likelihood (ML) classification was performed on the hazy datasets using training pixels extracted from the respective datasets. The accuracy of the classification was computed using confusion matrices where individual class and overall accuracy were determined. The results show that individual class accuracy is influenced not only by haze concentration but also class spectral properties. Overall classification accuracy declines with faster rate as visibility gets poorer.

The purpose of this study is to provide a systematic approach to address the industrial challenge of recognizing impeccable and defective gears by analyzing the spectrum of sound waves produced by the gears when in operation if attached to a testing circuit during sorting. This work classifies the gears into two classes, impeccable and defective classes. The spectra from several samples from both impeccable and defective gears were analyzed and five audio features were extracted from their spectra namely short-time energy, zero-crossing rate, Spectral entropy, pitch, and block energy entropy. It was found that there is a significant difference between the two classes. In training the algorithm, 5D features vectors from 20 feature vectors from impeccable gears and another 5D features vector from 20 defective gears as training samples to determine the discriminating point. In testing the algorithm, 20 samples were extracted randomly from the impeccable and defective gears but whose status was clearly known by visual inspection. The results of gear status given by the algorithm were compared to that of visual inspection. The samples were classified by using the Support Vector Machine (SVM) learning classification approach. A promising efficiency of 95% was obtained.

Investigation of acoustic and visual features for acoustic scene classification

Vehicle classification is one of the main focuses in traffic survey work. In this paper, an influenced recognition model of vehicle class was proposed and experimented on using the feature extraction of vehicle’s time-domain sound signal, considering the existing problems like costly equipments, low efficiency in operation, and difficulty of data analysis, etc. According to vehicle’s sound signal waveform characteristics, it is shown that the changed point can be obtained by analyzing the short-time energy and zero-crossing rate in time-domain, and then the vehicle class can be determined. The simulated experimental results indicate that the proposed method is simple and economical to recognize vehicle class. The statistic precision can correspond with the need of the engineering.

Epilepsy is a neurological disorder in which the brain is transiently altered. Predicting outcomes in epilepsy is essential for providing feedback that can foster improved outcomes in the future. This study aimed to investigate whether applying spectral and temporal filters to resting-state electroencephalography (EEG) signals could improve the prediction of outcomes for patients taking antiseizure medication to treat temporal lobe epilepsy (TLE). We collected EEG data from a total of 46 patients (divided into a seizure-free group (SF, n = 22) and a non-seizure-free group (NSF, n = 24)) with TLE and retrospectively reviewed their clinical data. We segmented spectral and temporal ranges with various time-domain features (Hjorth parameters, statistical parameters, energy, zero-crossing rate, inter-channel correlation, inter-channel phase locking value and spectral information derived from Fourier transform, Stockwell transform, and wavelet transform) and compared their performance by applying an optimal frequency strategy, an optimal duration strategy, and a combination strategy. For all time-domain features, the optimal frequency and time combination strategy showed the highest performance in distinguishing SF patients from NSF patients (area under the curve (AUC) = 0.790 ± 0.159). Furthermore, optimal performance was achieved by utilizing a feature vector derived from statistical parameters within the 39- to 41-Hz frequency band with a window length of 210 s, as evidenced by an AUC of 0.748. By identifying the optimal parameters, we improved the performance of the prediction model. These parameters can serve as standard parameters for predicting outcomes based on resting-state EEG signals.