Search results for: Speech Segmentation

Authors: Dejan Stantic, Jun Jo

Abstract:

There have been significant improvements in automatic voice recognition technology. However, existing systems still face difficulties, particularly when used by non-native speakers with accents. In this paper we address a problem of identifying the English accented speech of speakers from different backgrounds. Once an accent is identified the speech recognition software can utilise training set from appropriate accent and therefore improve the efficiency and accuracy of the speech recognition system. We introduced the Q factor, which is defined by the sum of relationships between frequencies of the formants. Four different accents were considered and experimented for this research. A scoring method was introduced in order to effectively analyse accents. The proposed concept indicates that the accent could be identified by analysing their formants.

Keywords: Accent Identification, Formants, Q Factor.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2061

Authors: Mary Jane B. Suarez

Abstract:

Speech anxiety is innumerable in any traditional communication classes especially for ESL students. The speech anxiety intensifies when communication skills assessments have taken its toll in an online mode of learning due to the perils of the COVID-19 virus. Teachers and students have experienced vast ambiguity on how to realize a still effective way to teach and learn various speaking skills amidst the pandemic. This mixed method study determined the factors that affected the public speaking skills of students in online performances, delineated the assessment gaps in assessing speaking skills in an online setup, and recommended ways to address students’ speech anxieties. Using convergent parallel design, quantitative data were gathered by examining the desired learning competencies of the English course including a review of the teacher’s class record to analyze how students’ performances reflected a significantly high level of anxiety in online speech delivery. Focus group discussion was also conducted for qualitative data describing students’ public speaking anxiety and assessment gaps. Results showed a significantly high level of students’ speech anxiety affected by time constraints, use of technology, lack of audience response, being conscious of making mistakes, and the use of English as a second language. The study presented recommendations to redesign curricular assessments of English teachers and to have a robust diagnosis of students’ speaking anxiety to better cater to the needs of learners in attempt to bridge any gaps in cultivating public speaking skills of students as educational institutions segue from the pandemic to the post-pandemic milieu.

Keywords: Blended learning, communication skills assessment, online speech delivery, public speaking anxiety, speech anxiety.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 140

Authors: Rosalyn R. Porle, Ali Chekima, Farrah Wong, G. Sainarayanan

Abstract:

Arms detection is one of the fundamental problems in human motion analysis application. The arms are considered as the most challenging body part to be detected since its pose and speed varies in image sequences. Moreover, the arms are usually occluded with other body parts such as the head and torso. In this paper, histogram-based skin colour segmentation is proposed to detect the arms in image sequences. Six different colour spaces namely RGB, rgb, HSI, TSL, SCT and CIELAB are evaluated to determine the best colour space for this segmentation procedure. The evaluation is divided into three categories, which are single colour component, colour without luminance and colour with luminance. The performance is measured using True Positive (TP) and True Negative (TN) on 250 images with manual ground truth. The best colour is selected based on the highest TN value followed by the highest TP value.

Keywords: image colour analysis, image motion analysis, skin, wavelet transform.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1537

Authors: Jong Han Joo, Jeong Hun Lee, Young Sun Kim, Jae Young Kang, Seung Ho Choi

Abstract:

In this study, we propose a novel technique for acoustic echo suppression (AES) during speech recognition under barge-in conditions. Conventional AES methods based on spectral subtraction apply fixed weights to the estimated echo path transfer function (EPTF) at the current signal segment and to the EPTF estimated until the previous time interval. However, the effects of echo path changes should be considered for eliminating the undesired echoes. We describe a new approach that adaptively updates weight parameters in response to abrupt changes in the acoustic environment due to background noises or double-talk. Furthermore, we devised a voice activity detector and an initial time-delay estimator for barge-in speech recognition in communication networks. The initial time delay is estimated using log-spectral distance measure, as well as cross-correlation coefficients. The experimental results show that the developed techniques can be successfully applied in barge-in speech recognition systems.

Keywords: Acoustic echo suppression, barge-in, speech recognition, echo path transfer function, initial delay estimator, voice activity detector.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2287

Authors: Yueming Wang, Rendong Ying, Sumxin Jiang, Peilin Liu

Abstract:

Unified Speech Audio Coding (USAC), the latest MPEG standardization for unified speech and audio coding, uses a speech/audio classification algorithm to distinguish speech and audio segments of the input signal. The quality of the recovered audio can be increased by well-designed orchestra/percussion classification and subsequent processing. However, owing to the shortcoming of the system, introducing an orchestra/percussion classification and modifying subsequent processing can enormously increase the quality of the recovered audio. This paper proposes an orchestra/percussion classification algorithm for the USAC system which only extracts 3 scales of Mel-Frequency Cepstral Coefficients (MFCCs) rather than traditional 13 scales of MFCCs and use Iterative Dichotomiser 3 (ID3) Decision Tree rather than other complex learning method, thus the proposed algorithm has lower computing complexity than most existing algorithms. Considering that frequent changing of attributes may lead to quality loss of the recovered audio signal, this paper also design a modified subsequent process to help the whole classification system reach an accurate rate as high as 97% which is comparable to classical 99%.

Keywords: ID3 Decision Tree, MFCC, Orchestra/Percussion Classification, USAC

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1653

Authors: Yong Zhang, Ruimin Hu

Abstract:

This paper presents a vocoder to obtain high quality synthetic speech at 600 bps. To reduce the bit rate, the algorithm is based on a sinusoidally excited linear prediction model which extracts few coding parameters, and three consecutive frames are grouped into a superframe and jointly vector quantization is used to obtain high coding efficiency. The inter-frame redundancy is exploited with distinct quantization schemes for different unvoiced/voiced frame combinations in the superframe. Experimental results show that the quality of the proposed coder is better than that of 2.4kbps LPC10e and achieves approximately the same as that of 2.4kbps MELP and with high robustness.

Keywords: Speech coding, Vector quantization, linear predicition, Mixed sinusoidal excitation

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2167

Authors: Rajkishore Prasad, Hiroshi Saruwatari, Kiyohiro Shikano

Abstract:

This paper describes Independent Component Analysis (ICA) based fixed-point algorithm for the blind separation of the convolutive mixture of speech, picked-up by a linear microphone array. The proposed algorithm extracts independent sources by non- Gaussianizing the Time-Frequency Series of Speech (TFSS) in a deflationary way. The degree of non-Gaussianization is measured by negentropy. The relative performances of algorithm under random initialization and Null beamformer (NBF) based initialization are studied. It has been found that an NBF based initial value gives speedy convergence as well as better separation performance

Keywords: Blind signal separation, independent component analysis, negentropy, convolutive mixture.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1747

Authors: Pavel Chmelar, Martin Dobrovolny

Abstract:

Laser projection or laser footprint detection is today widely used in many fields of robotics, measurement or electronics. The system accuracy strictly depends on precise laser footprint detection on target objects. This article deals with the laser line detection based on the RGB segmentation and the component labeling. As a measurement device was used the developed optical rangefinder. The optical rangefinder is equipped with vertical sweeping of the laser beam and high quality camera. This system was developed mainly for automatic exploration and mapping of unknown spaces. In the first section is presented a new detection algorithm. In the second section are presented measurements results. The measurements were performed in variable light conditions in interiors. The last part of the article present achieved results and their differences between day and night measurements.

Keywords: Automatic mapping, color segmentation, component labeling, distance measurement, laser line detection, vector map.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 3532

Authors: Meysam Mohamad pour, Fardad Farokhi

Abstract:

In this paper in consideration of each available techniques deficiencies for speech recognition, an advanced method is presented that-s able to classify speech signals with the high accuracy (98%) at the minimum time. In the presented method, first, the recorded signal is preprocessed that this section includes denoising with Mels Frequency Cepstral Analysis and feature extraction using discrete wavelet transform (DWT) coefficients; Then these features are fed to Multilayer Perceptron (MLP) network for classification. Finally, after training of neural network effective features are selected with UTA algorithm.

Keywords: Multilayer perceptron (MLP) neural network, Discrete Wavelet Transform (DWT) , Mels Scale Frequency Filter , UTA algorithm.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2327

Authors: Shannon Steinmetz, Ellen Gethner

Abstract:

In this paper a theoretical foundation is developed to segment, analyze and associate patterns within audio. We explore this on imagery via sonified audio applied to our segmentation framework. The approach involves a geodesic estimator within the statistical manifold, parameterized by musical centricity. We demonstrate viability by processing a database of random imagery to produce statistically significant clusters of similar imagery content.

Keywords: Sonification, musical information geometry, image content extraction, automated quantification, audio segmentation, pattern recognition.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 372

Authors: Aree A. Mohammed

Abstract:

In this work we develop an object extraction method and propose efficient algorithms for object motion characterization. The set of proposed tools serves as a basis for development of objectbased functionalities for manipulation of video content. The estimators by different algorithms are compared in terms of quality and performance and tested on real video sequences. The proposed method will be useful for the latest standards of encoding and description of multimedia content – MPEG4 and MPEG7.

Keywords: Object extraction, Video indexing, Segmentation, Optical flow, Motion estimators.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1327

Authors: Jalal Karam, Raed Saad

Abstract:

This paper studies the effect of different compression constraints and schemes presented in a new and flexible paradigm to achieve high compression ratios and acceptable signal to noise ratios of Arabic speech signals. Compression parameters are computed for variable frame sizes of a level 5 to 7 Discrete Wavelet Transform (DWT) representation of the signals for different analyzing mother wavelet functions. Results are obtained and compared for Global threshold and level dependent threshold techniques. The results obtained also include comparisons with Signal to Noise Ratios, Peak Signal to Noise Ratios and Normalized Root Mean Square Error.

Keywords: Speech Compression, Wavelets.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1710

Authors: A. Sindhuja, V. Sadasivam

Abstract:

Breast Cancer is the most common malignancy in women and the second leading cause of death for women all over the world. Earlier the detection of cancer, better the treatment. The diagnosis and treatment of the cancer rely on segmentation of Sonoelastographic images. Texture features has not considered for Sonoelastographic segmentation. Sonoelastographic images of 15 patients containing both benign and malignant tumorsare considered for experimentation.The images are enhanced to remove noise in order to improve contrast and emphasize tumor boundary. It is then decomposed into sub-bands using single level Daubechies wavelets varying from single co-efficient to six coefficients. The Grey Level Co-occurrence Matrix (GLCM), Local Binary Pattern (LBP) features are extracted and then selected by ranking it using Sequential Floating Forward Selection (SFFS) technique from each sub-band. The resultant images undergo K-Means clustering and then few post-processing steps to remove the false spots. The tumor boundary is detected from the segmented image. It is proposed that Local Binary Pattern (LBP) from the vertical coefficients of Daubechies wavelet with two coefficients is best suited for segmentation of Sonoelastographic breast images among the wavelet members using one to six coefficients for decomposition. The results are also quantified with the help of an expert radiologist. The proposed work can be used for further diagnostic process to decide if the segmented tumor is benign or malignant.

Keywords: Breast Cancer, Segmentation, Sonoelastography, Tumor Detection.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2180

Authors: Chandrashekhar Bhat, Dilifa J. Noronha, Faber A. Saldanha

Abstract:

Transportation of long turbine blades from one place to another is a difficult process. Hence a feasibility study of modularization of wind turbine blade was taken from structural standpoint through finite element analysis. Initially, a non-segmented blade is modeled and its structural behavior is evaluated to serve as reference. The resonant, static bending and fatigue tests are simulated in accordance with IEC61400-23 standard for comparison purpose. The non-segmented test blade is separated at suitable location based on trade off studies and the segments are joined with an innovative double strap bonded joint configuration. The adhesive joint is modeled by adopting cohesive zone modeling approach in ANSYS. The developed blade model is analyzed for its structural response through simulation. Performances of both the blades are found to be similar, which indicates that, efficient segmentation of the long blade is possible which facilitates easy transportation of the blades and on site reassembling. The location selected for segmentation and adopted joint configuration has resulted in an efficient segmented blade model which proves the methodology adopted for segmentation was quite effective. The developed segmented blade appears to be the viable alternative considering its structural response specifically in fatigue within considered assumptions.

Keywords: Cohesive zone modeling, fatigue, segmentation, wind turbine blade.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 3268

Authors: G. Tamulevičius, A. Serackis, T. Sledevič, D. Navakauskas

Abstract:

We consider the biggest challenge in speech recognition – noise reduction. Traditionally detected transient noise pulses are removed with the corrupted speech using pulse models. In this paper we propose to cope with the problem directly in Dynamic Time Warping domain. Bidirectional Dynamic Time Warping algorithm for the recognition of isolated words impacted by transient noise pulses is proposed. It uses simple transient noise pulse detector, employs bidirectional computation of dynamic time warping and directly manipulates with warping results. Experimental investigation with several alternative solutions confirms effectiveness of the proposed algorithm in the reduction of impact of noise on recognition process – 3.9% increase of the noisy speech recognition is achieved.

Keywords: Transient noise pulses, noise reduction, dynamic time warping, speech recognition.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1899

Authors: Serkan Ozbay, Ergun Ercelebi

Abstract:

Automatic Vehicle Identification (AVI) has many applications in traffic systems (highway electronic toll collection, red light violation enforcement, border and customs checkpoints, etc.). License Plate Recognition is an effective form of AVI systems. In this study, a smart and simple algorithm is presented for vehicle-s license plate recognition system. The proposed algorithm consists of three major parts: Extraction of plate region, segmentation of characters and recognition of plate characters. For extracting the plate region, edge detection algorithms and smearing algorithms are used. In segmentation part, smearing algorithms, filtering and some morphological algorithms are used. And finally statistical based template matching is used for recognition of plate characters. The performance of the proposed algorithm has been tested on real images. Based on the experimental results, we noted that our algorithm shows superior performance in car license plate recognition.

Keywords: Character recognizer, license plate recognition, plate region extraction, segmentation, smearing, template matching.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 7561

Authors: C. Lodato, S. Lopes

Abstract:

This paper describes a segmentation algorithm based on the cooperation of an optical flow estimation method with edge detection and region growing procedures. The proposed method has been developed as a pre-processing stage to be used in methodologies and tools for video/image indexing and retrieval by content. The addressed problem consists in extracting whole objects from background for producing images of single complete objects from videos or photos. The extracted images are used for calculating the object visual features necessary for both indexing and retrieval processes. The first task of the algorithm exploits the cues from motion analysis for moving area detection. Objects and background are then refined using respectively edge detection and region growing procedures. These tasks are iteratively performed until objects and background are completely resolved. The developed method has been applied to a variety of indoor and outdoor scenes where objects of different type and shape are represented on variously textured background.

Keywords: Motion Detection, Object Extraction, Optical Flow, Segmentation.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1869

Authors: Mohamed Ali KAMMOUN, Ahmed Ben HAMIDA

Abstract:

In this paper, we present a novel statistical approach to corpus-based speech synthesis. Classically, phonetic information is defined and considered as acoustic reference to be respected. In this way, many studies were elaborated for acoustical unit classification. This type of classification allows separating units according to their symbolic characteristics. Indeed, target cost and concatenation cost were classically defined for unit selection. In Corpus-Based Speech Synthesis System, when using large text corpora, cost functions were limited to a juxtaposition of symbolic criteria and the acoustic information of units is not exploited in the definition of the target cost. In this manuscript, we token in our consideration the unit phonetic information corresponding to acoustic information. This would be realized by defining a probabilistic linguistic Bi-grams model basically used for unit selection. The selected units would be extracted from the English TIMIT corpora.

Keywords: Unit selection, Corpus-based Speech Synthesis, Bigram model

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1415

Authors: Seyyed Yasser Hashemi

Abstract:

Skew detection and correction (SDC) has a direct effect in efficiency and exactitude of documents’ segmentation and analysis and thus is considered as a very important step in documents’ analysis field. Skew is a major problem in documents’ analysis for every language. For Arabic/Persian document scripts this problem is more severe because of special features of these languages. In this paper an efficient and fast algorithm for Document Skew Detection (DSD) based on the concept of segmentation and Center of Gravity (COG) is proposed. This algorithm is examined for 150 Arabic/Persian and English documents and SDC process are done successfully for 93 percent of documents with error rate of less than 1°. This algorithm shows better results for English documents compared to Arabic/Persian documents. The proposed method is also represents favorable results for handwritten, printed and also complicated documents such as newspapers and journals even with very low quality and resolution.

Keywords: Arabic/Persian document, Baseline, Centre of gravity, Document segmentation, Skew detection and correction.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1880

Authors: Abdelouahab Moussaoui, Nabila Ferahta, Victor Chen

Abstract:

The development of aid's systems for the medical diagnosis is not easy thing because of presence of inhomogeneities in the MRI, the variability of the data from a sequence to the other as well as of other different source distortions that accentuate this difficulty. A new automatic, contextual, adaptive and robust segmentation procedure by MRI brain tissue classification is described in this article. A first phase consists in estimating the density of probability of the data by the Parzen-Rozenblatt method. The classification procedure is completely automatic and doesn't make any assumptions nor on the clusters number nor on the prototypes of these clusters since these last are detected in an automatic manner by an operator of mathematical morphology called skeleton by influence zones detection (SKIZ). The problem of initialization of the prototypes as well as their number is transformed in an optimization problem; in more the procedure is adaptive since it takes in consideration the contextual information presents in every voxel by an adaptive and robust non parametric model by the Markov fields (MF). The number of bad classifications is reduced by the use of the criteria of MPM minimization (Maximum Posterior Marginal).

Keywords: Clustering, Automatic Classification, SKIZ, MarkovFields, Image segmentation, Maximum Posterior Marginal (MPM).

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1388

Authors: Sangjun Park, Jungpyo Hong, Byung-Ok Kang, Yun-keun Lee, Minsoo Hahn

Abstract:

In this paper, an algorithm for detecting and attenuating puff noises frequently generated under the mobile environment is proposed. As a baseline system, puff detection system is designed based on Gaussian Mixture Model (GMM), and 39th Mel Frequency Cepstral Coefficient (MFCC) is extracted as feature parameters. To improve the detection performance, effective acoustic features for puff detection are proposed. In addition, detected puff intervals are attenuated by high-pass filtering. The speech recognition rate was measured for evaluation and confusion matrix and ROC curve are used to confirm the validity of the proposed system.

Keywords: Gaussian mixture model, puff detection and cancellation, speech enhancement.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2201

Authors: António J. R. Neves, Rui Garcia, Paulo Dias, Alina Trifan

Abstract:

With the aging of the world population and the continuous growth in technology, service robots are more and more explored nowadays as alternatives to healthcare givers or personal assistants for the elderly or disabled people. Any service robot should be capable of interacting with the human companion, receive commands, navigate through the environment, either known or unknown, and recognize objects. This paper proposes an approach for object recognition based on the use of depth information and color images for a service robot. We present a study on two of the most used methods for object detection, where 3D data is used to detect the position of objects to classify that are found on horizontal surfaces. Since most of the objects of interest accessible for service robots are on these surfaces, the proposed 3D segmentation reduces the processing time and simplifies the scene for object recognition. The first approach for object recognition is based on color histograms, while the second is based on the use of the SIFT and SURF feature descriptors. We present comparative experimental results obtained with a real service robot.

Keywords: Service Robot, Object Recognition, 3D Sensors, Plane Segmentation.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1645

Authors: Mohammad Shoyaib, M. Abdullah-Al-Wadud, Oksam Chae

Abstract:

In this paper an effective approach for segmenting human skin regions in images taken at different environment is proposed. The proposed method uses a color distance map that is flexible enough to reliably detect the skin regions even if the illumination conditions of the image vary. Local image conditions is also focused, which help the technique to adaptively detect differently illuminated skin regions of an image. Moreover, usage of local information also helps the skin detection process to get rid of picking up much noisy pixels.

Keywords: Color Distance map, Reference skin color, Regiongrowing, Skin segmentation.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1975

Authors: Sheena Christabel Pravin, M. Palanivelan

Abstract:

Speech disfluencies are common in spontaneous speech. The primary purpose of this study was to distinguish linguistic disfluencies from stuttering disfluencies in bilingual Tamil–English (TE) speaking children. The secondary purpose was to determine whether their disfluencies are mediated by native language dominance and/or on an early onset of developmental stuttering at childhood. A detailed study was carried out to identify the prosodic and acoustic features that uniquely represent the disfluent regions of speech. This paper focuses on statistical modeling of repetitions, prolongations, pauses and interjections in the speech corpus encompassing bilingual spontaneous utterances from school going children – English and Tamil. Two classifiers including Hidden Markov Models (HMM) and the Multilayer Perceptron (MLP), which is a class of feed-forward artificial neural network, were compared in the classification of disfluencies. The results of the classifiers document the patterns of disfluency in spontaneous speech samples of school-aged children to distinguish between Children Who Stutter (CWS) and Children with Language Impairment CLI). The ability of the models in classifying the disfluencies was measured in terms of F-measure, Recall, and Precision.

Keywords: Bilingual, children who stutter, children with language impairment, Hidden Markov Models, multi-layer perceptron, linguistic disfluencies, stuttering disfluencies.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1002

Authors: Wernhuar Tarng, Yuan-Yuan Chen, Chien-Lung Li, Kun-Rong Hsie, Mingteh Chen

Abstract:

An emotional speech recognition system for the applications on smart phones was proposed in this study to combine with 3G mobile communications and social networks to provide users and their groups with more interaction and care. This study developed a mechanism using the support vector machines (SVM) to recognize the emotions of speech such as happiness, anger, sadness and normal. The mechanism uses a hierarchical classifier to adjust the weights of acoustic features and divides various parameters into the categories of energy and frequency for training. In this study, 28 commonly used acoustic features including pitch and volume were proposed for training. In addition, a time-frequency parameter obtained by continuous wavelet transforms was also used to identify the accent and intonation in a sentence during the recognition process. The Berlin Database of Emotional Speech was used by dividing the speech into male and female data sets for training. According to the experimental results, the accuracies of male and female test sets were increased by 4.6% and 5.2% respectively after using the time-frequency parameter for classifying happy and angry emotions. For the classification of all emotions, the average accuracy, including male and female data, was 63.5% for the test set and 90.9% for the whole data set.

Keywords: Smart phones, emotional speech recognition, socialnetworks, support vector machines, time-frequency parameter, Mel-scale frequency cepstral coefficients (MFCC).

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1822

Authors: L. Salhi, M. Talbi, A. Cherif

Abstract:

This paper presents a new strategy of identification and classification of pathological voices using the hybrid method based on wavelet transform and neural networks. After speech acquisition from a patient, the speech signal is analysed in order to extract the acoustic parameters such as the pitch, the formants, Jitter, and shimmer. Obtained results will be compared to those normal and standard values thanks to a programmable database. Sounds are collected from normal people and patients, and then classified into two different categories. Speech data base is consists of several pathological and normal voices collected from the national hospital “Rabta-Tunis". Speech processing algorithm is conducted in a supervised mode for discrimination of normal and pathology voices and then for classification between neural and vocal pathologies (Parkinson, Alzheimer, laryngeal, dyslexia...). Several simulation results will be presented in function of the disease and will be compared with the clinical diagnosis in order to have an objective evaluation of the developed tool.

Keywords: Formants, Neural Networks, Pathological Voices, Pitch, Wavelet Transform.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2817

Authors: Tsvetomira Braynova, Miglena Simonska

Abstract:

The article presents the results of a research of phonological processes in three- and four-year-old children. A test, created for the purpose of the study, was developed and conducted among 120 children. The study included three areas of research - at the level of words (96 words), at the level of sentence repetition (10 sentences) and at the level of generating own speech from a picture (15 pictures). The test also gives us additional information about the articulation errors of the assessed children. The main purpose of the research is to analyze all phonological processes that occur at this age in Bulgarian children and to identify which are typical and atypical for this age. The results show that the most common phonology errors that children make are: sound substitution, elision of sound, metathesis of sound, elision of syllable, elision of consonants clustered in a syllable. Measuring the correlation between average length of repeated speech and average length of generated speech, the analysis does not prove that the more words a child can repeat in part “repeated speech”, the more words they can be expected to generate in part “generating sentence”. The results of this study show that the task of naming a word provides sufficient and representative information to assess the child's phonology.

Keywords: Articulation, phonology, speech, language development.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 337

Authors: Kamil R. Aida-zade, Jamaladdin Z. Hasanov

Abstract:

An approach is offered for more precise definition of base lines- borders in handwritten cursive text and general problems of handwritten text segmentation have also been analyzed. An offered method tries to solve problems arose in handwritten recognition with specific slant or in other words, where the letters of the words are not on the same vertical line. As an informative features, some recognition systems use ascending and descending parts of the letters, found after the word-s baseline detection. In such recognition systems, problems in baseline detection, impacts the quality of the recognition and decreases the rate of the recognition. Despite other methods, here borders are found by small pieces containing segmentation elements and defined as a set of linear functions. In this method, separate borders for top and bottom border lines are found. At the end of the paper, as a result, azerbaijani cursive handwritten texts written in Latin alphabet by different authors has been analyzed.

Keywords: Azeri, azerbaijani, latin, segmentation, cursive, HWR, handwritten, recognition, baseline, ascender, descender, symbols.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2458

Authors: Y. Deng, X. Li, C. Kwan, B. Raj, R. Stern

Abstract:

The current speech interfaces in many military applications may be adequate for native speakers. However, the recognition rate drops quite a lot for non-native speakers (people with foreign accents). This is mainly because the nonnative speakers have large temporal and intra-phoneme variations when they pronounce the same words. This problem is also complicated by the presence of large environmental noise such as tank noise, helicopter noise, etc. In this paper, we proposed a novel continuous acoustic feature adaptation algorithm for on-line accent and environmental adaptation. Implemented by incremental singular value decomposition (SVD), the algorithm captures local acoustic variation and runs in real-time. This feature-based adaptation method is then integrated with conventional model-based maximum likelihood linear regression (MLLR) algorithm. Extensive experiments have been performed on the NATO non-native speech corpus with baseline acoustic model trained on native American English. The proposed feature-based adaptation algorithm improved the average recognition accuracy by 15%, while the MLLR model based adaptation achieved 11% improvement. The corresponding word error rate (WER) reduction was 25.8% and 2.73%, as compared to that without adaptation. The combined adaptation achieved overall recognition accuracy improvement of 29.5%, and WER reduction of 31.8%, as compared to that without adaptation.

Keywords: speaker adaptation; environment adaptation; robust speech recognition; SVD; non-native speech recognition

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 3194

Authors: Sungju Yun, Anjin Park, Keechul Jung

Abstract:

Motion capture devices have been utilized in producing several contents, such as movies and video games. However, since motion capture devices are expensive and inconvenient to use, motions segmented from captured data was recycled and synthesized to utilize it in another contents, but the motions were generally segmented by contents producers in manual. Therefore, automatic motion segmentation is recently getting a lot of attentions. Previous approaches are divided into on-line and off-line, where on-line approaches segment motions based on similarities between neighboring frames and off-line approaches segment motions by capturing the global characteristics in feature space. In this paper, we propose a graph-based high-level motion segmentation method. Since high-level motions consist of several repeated frames within temporal distances, we consider all similarities among all frames within the temporal distance. This is achieved by constructing a graph, where each vertex represents a frame and the edges between the frames are weighted by their similarity. Then, normalized cuts algorithm is used to partition the constructed graph into several sub-graphs by globally finding minimum cuts. In the experiments, the results using the proposed method showed better performance than PCA-based method in on-line and GMM-based method in off-line, as the proposed method globally segment motions from the graph constructed based similarities between neighboring frames as well as similarities among all frames within temporal distances.

Keywords: Capture Devices, High-Level Motion, Motion Segmentation, Normalized Cuts

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1290