Search
Filters
total: 76
Search results for: automated pronunciation error detection
-
Automated detection of pronunciation errors in non-native English speech employing deep learning
PublicationDespite significant advances in recent years, the existing Computer-Assisted Pronunciation Training (CAPT) methods detect pronunciation errors with a relatively low accuracy (precision of 60% at 40%-80% recall). This Ph.D. work proposes novel deep learning methods for detecting pronunciation errors in non-native (L2) English speech, outperforming the state-of-the-art method in AUC metric (Area under the Curve) by 41%, i.e., from...
-
Weakly-Supervised Word-Level Pronunciation Error Detection in Non-Native English Speech
PublicationWe propose a weakly-supervised model for word-level mispronunciation detection in non-native (L2) English speech. To train this model, phonetically transcribed L2 speech is not required and we only need to mark mispronounced words. The lack of phonetic transcriptions for L2 speech means that the model has to learn only from a weak signal of word-level mispronunciations. Because of that and due to the limited amount of mispronounced...
-
Computer-assisted pronunciation training—Speech synthesis is almost all you need
PublicationThe research community has long studied computer-assisted pronunciation training (CAPT) methods in non-native speech. Researchers focused on studying various model architectures, such as Bayesian networks and deep learning methods, as well as on the analysis of different representations of the speech signal. Despite significant progress in recent years, existing CAPT methods are not able to detect pronunciation errors with high...
-
Evaluation of aspiration problems in L2 English pronunciation employing machine learning
PublicationThe approach proposed in this study includes methods specifically dedicated to the detection of allophonic variation in English. This study aims to find an efficient method for automatic evaluation of aspiration in the case of Polish second-language (L2) English speakers’ pronunciation when whole words are analyzed instead of particular allophones extracted from words. Sample words including aspirated and unaspirated allophones...
-
Automated Reduced Model Order Selection
PublicationThis letter proposes to automate generation of reduced-order models used for accelerated -parameter computation by applying a posteriori model error estimators. So far,a posteriori error estimators were used in Reduced Basis Method (RBM) and Proper Orthogonal Decomposition (POD) to select frequency points at which basis vectors are generated. This letter shows how a posteriori error estimators can be applied to automatically select...
-
Greedy Multipoint Model-Order Reduction Technique for Fast Computation of Scattering Parameters of Electromagnetic Systems
PublicationThis paper attempts to develop a new automated multipoint model-order reduction (MOR) technique, based on matching moments of the system input–output function, which would be suited for fast and accurate computation of scattering parameters for electromagnetic (EM) systems over a wide frequency band. To this end, two questions are addressed. Firstly, the cost of the wideband reduced model generation is optimized by automating a...
-
Automatic Marking of Allophone Boundaries in Isolated English spoken Words
PublicationThe work presents a method that allows delimiting the borders of allophones in isolated English words. The described method is based on the DTW algorithm combining two signals, a reference signal and an analyzed one. As the reference signal, recordings from the MODALITY database were used, from which the words were extracted. This database was also used for tests, which were described. Test results show that the automatic determination...
-
Parametrized Local Reduced-Order Models With Compressed Projection Basis for Fast Parameter-Dependent Finite-Element Analysis
PublicationThis paper proposes an automated parametric local model-order reduction scheme for the expedited design of microwave devices using the full-wave finite-element method (FEM). The approach proposed here results in parameterized reduced-order models (ROMs) that account for the geometry and material variation in the selected subregion of the structure. In each subregion, a parameter-dependent projection basis is generated by concatenating...
-
A Development of a Capacitive Voltage Divider for High Voltage Measurement as Part of a Combined Current and Voltage Sensor
PublicationThis article deals with the development of capacitive voltage divider for high voltage measurements and presents a method of analysis and optimization of its parameters. This divider is a part of a combined voltage and current sensor for measurements in high voltage power networks. The sensor allows continuous monitoring of the network distribution status and performs a quick diagnosis and location of possible network failures....
-
3D Monitoring - Identification of measurement problems at larger movements of the tracked points
PublicationAuthors identified the problems associated with the determination of the controlled points coordinates by use of automated Total Station placed behind transparent barrier. Important thing in the mentioned analysis was a large change of controlled points position and not stable Total Station’s stand (because of stand’s thermal drift). This two elements, combined with measurement made through glass plate determine the need for impact...
-
Artificial intelligence for software development — the present and the challenges for the future
PublicationSince the time when first CASE (Computer-Aided Software Engineering) methods and tools were developed, little has been done in the area of automated creation of code. CASE tools support a software engineer in creation the system structure, in defining interfaces and relationships between software modules and, after the code has been written, in performing testing tasks on different levels of detail. Writing code is still the task...
-
Assessment of the Steering Precision of a Hydrographic Unmanned Surface Vessel (USV) along Sounding Profiles Using a Low-Cost Multi-Global Navigation Satellite System (GNSS) Receiver Supported Autopilot
Publicationhe performance of bathymetric measurements by traditional methods (using manned vessels) in ultra-shallow waters, i.e., lakes, rivers, and sea beaches with a depth of less than 1 m, is often difficult or, in many cases, impossible due to problems related to safe vessel maneuvering. For this reason, the use of shallow draft hydrographic Unmanned Surface Vessels (USV) appears to provide a promising alternative method for performing...
-
Octave Error Immune and Instantaneous Pitch Detection Algorithm.
PublicationCelem publikacji jest prezentacja odpornego na błędy oktawowe, bazującego na analizie widmowej algorytmu detekcji częstotliwości podstawowej. Zaproponowana metoda dobrze sobie radzi z sygnałami o dużej zawartości sygnałów harmonicznych, jak i z prawie sinusoidalnymi przebiegami. Eksperymenty przeprowadzonno na 567 dzwiękach instrumentów muzycznych. Dźwięki grane były z różnymi artykulacjami, dynamiką i reprezentowałe były w całej...
-
Fully Automated AI-powered Contactless Cough Detection based on Pixel Value Dynamics Occurring within Facial Regions
PublicationIncreased interest in non-contact evaluation of the health state has led to higher expectations for delivering automated and reliable solutions that can be conveniently used during daily activities. Although some solutions for cough detection exist, they suffer from a series of limitations. Some of them rely on gesture or body pose recognition, which might not be possible in cases of occlusions, closer camera distances or impediments...
-
High accuracy and octave error immune pitch detection algorithms.
PublicationW publikacji przedstawiona została metoda poprawiająca dokładność estymacji częstotliwości podstawowej dźwięków naturalnych i syntetycznych. Opracowany algorytm wykorzystuje sztczną sieć neuronową. Dodatkowo przedstawiony został algorytm zoptymalizowany pod kątem błędów oktawowych, operujący w dziedzinie częstotliwości. Przedstawiona metoda jest bardzo skuteczna zarówno dla sygnałów harmonicznych o znaczącej energii poszczególnych...
-
Problemy zarządzania bezpieczeństwem obiektu przemysłowego podwyższonego ryzyka
PublicationW rozdziale przedstawiono wybrane zagadnienia dotyczące zarządzania bezpieczeństwem w zautomatyzowanym złożonym obiekcie podwyższonego ryzyka. Pokazano, że ryzyko strat można istotnie ograniczyć stosując odpowiednie rozwiązania techniczne w postaci warstwowego systemu zabezpieczeń, który obejmuje podstawowy system sterowania procesem, człowieka-operatora i system automatyki zabezpieczeniowej. Podkreślono znaczenie właściwego zaprojektowania...
-
Accurate Modeling of Frequency Selective Surfaces Using Fully-Connected Regression Model with Automated Architecture Determination and Parameter Selection Based on Bayesian Optimization
PublicationSurrogate modeling has become an important tool in the design of high-frequency structures. Although full-wave electromagnetic (EM) simulation tools provide an accurate account for the circuit characteristics and performance, they entail considerable computational expenditures. Replacing EM analysis by fast surrogates provides a way to accelerate the design procedures. Unfortunately, modeling of microwave passives is a challenging...
-
An adaptive-noise Augmented Kalman Filter approach for input-state estimation in structural dynamics
PublicationThe establishment of a Digital Twin of an operating engineered system can increase the potency of Structural Health Monitoring (SHM) tools, which are then bestowed with enhanced predictive capabilities. This is particularly relevant for wind energy infrastructures, where the definition of remaining useful life is a main driver for assessing the efficacy of these systems. In order to ensure a proper representation of the physical...
-
Automated Detection of Sleep Apnea and Hypopnea Events Based on Robust Airflow Envelope Tracking in the Presence of Breathing Artifacts. - [IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS]
PublicationThe paper presents a new approach to detection of apnea/hypopnea events, in the presence of artifacts and breathing irregularities, from a single channel airflow record. The proposed algorithm, based on a robust envelope detector , identifies segments of signal affected by a high amplitude mo d- ulation corresponding to apnea/hypopnea events. It is show n that a robust airflow envelope - free of breathing artifacts - improves effectiveness...
-
MiMSeg - an algorithm for automated detection of tumor tissue on NMR apparent diffusion coefficient maps.
Publication -
Automated detection of sleep apnea and hypopnea events based on robust airflow envelope tracking
PublicationThe paper presents a new approach to detection of apnea/hypopnea events, in the presence of artifacts and breathing irregularities, from a single-channel airflow record. The proposed algorithm identifies segments of signal affected by a high amplitude modulation corresponding to apnea/hypopnea events. It is shown that a robust airflow envelope—free of breathing artifacts—improves effectiveness of the diagnostic process and allows...
-
A Novel Divisive iK-Means Algorithm with Region-Driven Feature Selection as a Tool for Automated Detection of Tumour Heterogeneity in MALDI IMS Experiments
Publication -
Automated Diagnostics of Current Pick-Up Disturbances in Electric Traction Networks
PublicationThe present work defines the basic causes of bow disturbances of current pick-up, sets a task of establishing a system of automated control of bow disturbances at feeder zones of electric traction networks, proposes structural variants of the technical system implementation, describes the algorithm of detection of bow disturbances of current pick-up.
-
Noise effect on parameters of quiet sonar with code modulation
PublicationEarlier publications of the paper authors have shown that the use of code keying mixed with the CW FM sound signal allows the significant reduction in the distance measurement error, compared to classic silent CW FM sonar. In addition to the code modulation parameters, the magnitude of this error is influenced by the received input acoustic noise. The article shows the dependence of the input signal-to-noise ratio and the sound...
-
Mispronunciation Detection in Non-Native (L2) English with Uncertainty Modeling
PublicationA common approach to the automatic detection of mispronunciation in language learning is to recognize the phonemes produced by a student and compare it to the expected pronunciation of a native speaker. This approach makes two simplifying assumptions: a) phonemes can be recognized from speech with high accuracy, b) there is a single correct way for a sentence to be pronounced. These assumptions do not always hold, which can result...
-
PCR-ELISA: inexpensive alternative to quantitative PCR
PublicationPCR-ELISA (polymerase chain reaction-enzyme linked immunosorbent assay), a combination of PCR and ELISA methods, has been used since late 1980s. The technique is based on specially labelled DNA fragments which are captured by specific DNA sequences and detected by antibodies. The whole procedure of PCR-ELISA is divided into three steps: DNA extraction, PCR reaction and detection by ELISA. The method has been found as very specific...
-
Respiratory Rate Estimation Based on Detected Mask Area in Thermal Images
PublicationThe popularity of non-contact methods of measuring vital signs, particularly respiratory rate, has increased during the SARS-COV-2 pandemic. Breathing parameters can be estimated by analysis of temperature changes observed in thermal images of nostrils or mouth regions. However, wearing virus-protection face masks prevents direct detection of such face regions. In this work, we propose to use an automatic mask detection approach...
-
Hierarchical 2-step neural-based LEGO bricks detection and labeling
PublicationLEGO bricks are extremely popular and allow the creation of almost any type of construction due to multiple shapes available. LEGO building requires however proper brick arrangement, usually done by shape. With over 3700 different LEGO parts this can be troublesome. In this paper, we propose a solution for object detection and annotation on images. The solution is designed as a part of an automated LEGO bricks arrangement. The...
-
Feature Reduction Using Similarity Measure in Object Detector Learning with Haar-like Features
PublicationThis paper presents two methods of training complexity reduction by additional selection of features to check in object detector training task by AdaBoost training algorithm. In the first method, the features with weak performance at first weak classifier building process are reduced based on a list of features sorted by minimum weighted error. In the second method the feature similarity measures are used to throw away that features...
-
Computer-aided analysis of signals from a low-coherence Fabry-Perot interferometer used for measurements of biological samples
PublicationThe aim of the study was to develop an automated computer-aided system for analysis of spectrograms obtained from measurements of biological samples performed with a low-coherence Fabry-Pérot interferometer. Information necessary to determine dispersion characteristics of measured materials can be calculated from the positions of the maxima and minima that are present in their spectra. The main challenge faced during the development...
-
Comparison of Acoustic and Visual Voice Activity Detection for Noisy Speech Recognition
PublicationThe problem of accurate differentiating between the speaker utterance and the noise parts in a speech signal is considered. The influence of utilizing a voice activity detection in speech signals on the accuracy of the automatic speech recognition (ASR) system is presented. The examined methods of voice activity detection are based on acoustic and visual modalities. The problem of detecting the voice activity in clean and noisy...
-
Detection of Alzheimer's disease using Otsu thresholding with tunicate swarm algorithm and deep belief network
PublicationIntroduction: Alzheimer’s Disease (AD) is a degenerative brain disorder characterized by cognitive and memory dysfunctions. The early detection of AD is necessary to reduce the mortality rate through slowing down its progression. The prevention and detection of AD is the emerging research topic for many researchers. The structural Magnetic Resonance Imaging (sMRI) is an extensively used imaging technique in detection of AD, because...
-
Dekodowanie kodów iterowanych z użyciem sieci neuronowej
PublicationNadmiarowe kody iterowane są jedną z prostych metod pozyskiwania długich kodów korekcyjnych zapewniających dużą ochronę przed błędami. Jednocześnie, chociaż ich podstawowy iteracyjny dekoder jest prosty koncepcyjnie oraz łatwy w implementacji, to nie jest on rozwiązaniem optymalnym. Poszukując alternatywnych rozwiązań zaproponowano, przedstawioną w pracy, strukturę dekodera tego typu kodów wspomaganą przez sieci neuronowe. Zaproponowane...
-
DIAGNOSIS OF MALIGNANT MELANOMA BY NEURAL NETWORK ENSEMBLE-BASED SYSTEM UTILISING HAND-CRAFTED SKIN LESION FEATURES
PublicationMalignant melanomas are the most deadly type of skin cancer but detected early have high chances for successful treatment. In the last twenty years, the interest of automated melanoma recognition detection and classification dynamically increased partially because of public datasets appearing with dermatoscopic images of skin lesions. Automated computer-aided skin cancer detection in dermatoscopic images is a very challenging task...
-
Detection and Direction-of-Arrival Estimation of Weak Spread Spectrum Signals Received with Antenna Array
PublicationThis paper presents a method for the joint detection and direction of arrival (DOA) estimation of low probability of detection (LPD) signals. The proposed approach is based on using the antenna array to receive spread-spectrum signals hidden below the noise floor. Array processing exploits the spatial correlation between phase-delayed copies of the signal and allows us to evaluate the parameter used to make the decision about the...
-
Weighted Clustering for Bees Detection on Video Images
PublicationThis work describes a bee detection system to monitor bee colony conditions. The detection process on video images has been divided into 3 stages: determining the regions of interest (ROI) for a given frame, scanning the frame in ROI areas using the DNN-CNN classifier, in order to obtain a confidence of bee occurrence in each window in any position and any scale, and form one detection window from a cloud of windows provided by...
-
Visually validated semi-automatic high-frequency oscillation detection aides the delineation of epileptogenic regions during intra-operative electrocorticography
PublicationOBJECTIVE: To test the utility of a novel semi-automated method for detecting, validating, and quantifying high-frequency oscillations (HFOs): ripples (80-200 Hz) and fast ripples (200-600 Hz) in intra-operative electrocorticography (ECoG) recordings. METHODS: Sixteen adult patients with temporal lobe epilepsy (TLE) had intra-operative ECoG recordings at the time of resection. The computer-annotated ECoG recordings were visually...
-
Methodology for the Correction of the Spatial Orientation Angles of the Unmanned Aerial Vehicle Using Real Time GNSS, a Shoreline Image and an Electronic Navigational Chart
PublicationUndoubtedly, Low-Altitude Unmanned Aerial Vehicles (UAVs) are becoming more common in marine applications. Equipped with a Global Navigation Satellite System (GNSS) Real-Time Kinematic (RTK) receiver for highly accurate positioning, they perform camera and Light Detection and Ranging (LiDAR) measurements. Unfortunately, these measurements may still be subject to large errors-mainly due to the inaccuracy of measurement of the optical...
-
Playback detection using machine learning with spectrogram features approach
PublicationThis paper presents 2D image processing approach to playback detection in automatic speaker verification (ASV) systems using spectrograms as speech signal representation. Three feature extraction and classification methods: histograms of oriented gradients (HOG) with support vector machines (SVM), HAAR wavelets with AdaBoost classifier and deep convolutional neural networks (CNN) were compared on different data partitions in respect...
-
Behavioral state classification in epileptic brain using intracranial electrophysiology
PublicationOBJECTIVE: Automated behavioral state classification can benefit next generation implantable epilepsy devices. In this study we explored the feasibility of automated awake (AW) and slow wave sleep (SWS) classification using wide bandwidth intracranial EEG (iEEG) in patients undergoing evaluation for epilepsy surgery. APPROACH: Data from seven patients (age [Formula: see text], 4 women) who underwent intracranial depth electrode...
-
Methodology for Performing Bathymetric Measurements of Shallow Waterbodies Using an UAV, and their Processing Based on the SVR Algorithm
PublicationState-of-art methods of bathymetric measurements for shallow waterbodies use Global Navigation Satellite System (GNSS) receiver, bathymetric Light Detection and Ranging (LiDAR) sensor or satellite imagery. Currently, photogrammetric methods with the application of Unmanned Aerial Vehicles (UAV) are gathering great importance. This publication aims to present step-by-step methodology for carrying out the bathymetric measurements...
-
A New Direct-Sequence Spread Spectrum Signal Detection Method for Underwater Acoustic Communications in Shallow-Water Channel
PublicationDirect-Sequence Spread Spectrum (DSSS) is one of the modulation and coding techniques used in Underwater Acoustic Communication (UAC) systems for reliable data transmision even at low signal levels. However, in a shallow water channel, there is a strong multipath propagation which causes a phase fluctuation of the received signal, affecting the performance of the spread-spectrum system. The article presents a differential method...
-
Raman Spectra Measurements for Chemical Identifications - Aspect of Uncertainty Sources and Reduction of Their Effects
PublicationRaman spectrometers enable fast and non-contact identification of examined chemicals. These devices measure Raman spectra and compare with the spectra database to identify unknown and often illicit chemicals (e.g. drugs, explosives) usually without any sample preparation. Raman spectra measurements are a challenge due to noise and interferences present outside the laboratories (field applications). The design of a portable Raman...
-
Continuous wave sonar with hyperbolic frequency modulation keyed by pseudo-random sequence
PublicationA CW FM type sounding signal is used in the classical solution of silent sonar. While the signal provides a relatively simple implementation of digital signal processing, and ensures good detection conditions, unfortunately, in the presence of the Doppler effect, distance measurement results tend to be wrong. This is due to the fact that the received signal’s instantaneous frequency value is dependent both on the distance to the...
-
AN ALGORITHM FOR PORTAL HYPERTENSIVE GASTROPATHY RECOGNITION ON THE ENDOSCOPIC RECORDINGS
PublicationSymptoms recognition of portal hypertensive gastropathy (PHG) can be done by analysing endoscopic recordings, but manual analysis done by physician may take a long time. This increases probability of missing some symptoms and automated methods may be applied to prevent that. In this paper a novel hybrid algorithm for recognition of early stage of portal hypertensive gastropathy is proposed. First image preprocessing is described....
-
Cichy sonar - stan aktualny i perspektywy
PublicationW zastosowaniach militarnych często istnieje potrzeba prowadzenia obserwacji w sposób skryty za pomocą urządzeń emitujących sygnały trudne do przechwycenia przez przeciwnika. Rozwiązania stosowane w cichych radarach stanowiły punkt wyjścia do opracowania cichego sonaru. Prace nad projektem rozpoczęto na Politechnice Gdańskiej w roku 2010 w ramach Grantu NCBiR i są one dalej kontynuowane. W celu zachowania w cichym sonarze warunków...
-
Visual Lip Contour Detection for the Purpose of Speech Recognition
PublicationA method for visual detection of lip contours in frontal recordings of speakers is described and evaluated. The purpose of the method is to facilitate speech recognition with visual features extracted from a mouth region. Different Active Appearance Models are employed for finding lips in video frames and for lip shape and texture statistical description. Search initialization procedure is proposed and error measure values are...
-
Algorithms of chemicals detection using raman spectra
PublicationRaman spectrometers are devices which enable fast and non-contact identification of examined chemicals. These devices utilize the Raman phenomenon to identify unknown and often illicit chemicals (e.g. drugs, explosives)without the necessity of their preparation. Now, Raman devices can be portable and therefore can be more widely used to improve security at public places. Unfortunately, Raman spectra measurements is a challenge...
-
Rediscovering Automatic Detection of Stuttering and Its Subclasses through Machine Learning—The Impact of Changing Deep Model Architecture and Amount of Data in the Training Set
PublicationThis work deals with automatically detecting stuttering and its subclasses. An effective classification of stuttering along with its subclasses could find wide application in determining the severity of stuttering by speech therapists, preliminary patient diagnosis, and enabling communication with the previously mentioned voice assistants. The first part of this work provides an overview of examples of classical and deep learning...
-
Condition-Based Monitoring of DC Motors Performed with Autoencoders
PublicationThis paper describes a condition-based monitoring system estimating DC motor degradation with the use of an autoencoder. Two methods of training the autoencoder are evaluated, namely backpropagation and extreme learning machines. The root mean square (RMS) error in the reconstruction of successive fragments of the measured DC motor angular-frequency signal, which is fed to the input of autoencoder, is used to determine the health...