Speaker diarization: Segmentation and clustering of speeches

Speaker diarization is the process of sorting speeches according to the speaker. Diarization helps to search and retrieve what a certain speaker uttered in a meeting. Applications of diarization systemsextend to other domains than meetings, for example, lectures, telephone, television, and radio. Besides, diarization enhances the performance of several speech technologies such as speaker recognition, automatic transcription, and speaker tracking. Methodologies previously used in developing diarization systems are discussed. Prior results and techniques are studied and compared. Methods such as Hidden Markov Models and Gaussian Mixture Models that are used in speaker recognition and other speech technologies are also used in speaker diarization. The objective of this thesis is to develop a speaker diarization system in meeting domain. Experimental part of this work indicates that zero-crossing rate can be used effectively in breaking down the audio stream into segments, and adaptive Gaussian Models fit adequately short audio segments. Results show that 35 Gaussian Models and one second as average length of each segment are optimum values to build a diarization system for the tested data. Uniting the segments which are uttered by same speaker is done in a bottom-up clustering by a newapproach of categorizing the mixture weights.

Development of a Biometric Personal Authentication System Based on Fingerprint and Speech

Biometrics deals with the physiological and behavioral characteristics of an individual to establish identity. Fingerprint based authentication is the most advanced biometric authentication technology. The minutiae based fingerprint identification method offer reasonable identification rate. The feature minutiae map consists of about 70-100 minutia points and matching accuracy is dropping down while the size of database is growing up. Hence it is inevitable to make the size of the fingerprint feature code to be as smaller as possible so that identification may be much easier. In this research, a novel global singularity based fingerprint representation is proposed. Fingerprint baseline, which is the line between distal and intermediate phalangeal joint line in the fingerprint, is taken as the reference line. A polygon is formed with the singularities and the fingerprint baseline. The feature vectors are the polygonal angle, sides, area, type and the ridge counts in between the singularities. 100% recognition rate is achieved in this method. The method is compared with the conventional minutiae based recognition method in terms of computation time, receiver operator characteristics (ROC) and the feature vector length. Speech is a behavioural biometric modality and can be used for identification of a speaker. In this work, MFCC of text dependant speeches are computed and clustered using k-means algorithm. A backpropagation based Artificial Neural Network is trained to identify the clustered speech code. The performance of the neural network classifier is compared with the VQ based Euclidean minimum classifier. Biometric systems that use a single modality are usually affected by problems like noisy sensor data, non-universality and/or lack of distinctiveness of the biometric trait, unacceptable error rates, and spoof attacks. Multifinger feature level fusion based fingerprint recognition is developed and the performances are measured in terms of the ROC curve. Score level fusion of fingerprint and speech based recognition system is done and 100% accuracy is achieved for a considerable range of matching threshold

Speaker identification using models for phonemes

Motivation for Speaker recognition work is presented in the first part of the thesis. An exhaustive survey of past work in this field is also presented. A low cost system not including complex computation has been chosen for implementation. Towards achieving this a PC based system is designed and developed. A front end analog to digital convertor (12 bit) is built and interfaced to a PC. Software to control the ADC and to perform various analytical functions including feature vector evaluation is developed. It is shown that a fixed set of phrases incorporating evenly balanced phonemes is aptly suited for the speaker recognition work at hand. A set of phrases are chosen for recognition. Two new methods are adopted for the feature evaluation. Some new measurements involving a symmetry check method for pitch period detection and ACE‘ are used as featured. Arguments are provided to show the need for a new model for speech production. Starting from heuristic, a knowledge based (KB) speech production model is presented. In this model, a KB provides impulses to a voice producing mechanism and constant correction is applied via a feedback path. It is this correction that differs from speaker to speaker. Methods of defining measurable parameters for use as features are described. Algorithms for speaker recognition are developed and implemented. Two methods are presented. The first is based on the model postulated. Here the entropy on the utterance of a phoneme is evaluated. The transitions of voiced regions are used as speaker dependent features. The second method presented uses features found in other works, but evaluated differently. A knock—out scheme is used to provide the weightage values for the selection of features. Results of implementation are presented which show on an average of 80% recognition. It is also shown that if there are long gaps between sessions, the performance deteriorates and is speaker dependent. Cross recognition percentages are also presented and this in the worst case rises to 30% while the best case is 0%. Suggestions for further work are given in the concluding chapter.

Desenvolvimento de aplicativos usando reconhecimento e síntese de voz

A fala é um mecanismo natural para a interação homem-máquina. A tecnologia de processamento de fala (ou voz) encontra-se bastante avançada e, em escala mundial, existe vasta disponibilidade de software, tanto comercial quanto acadêmico. a maioria assume a disponibilidade de um reconhecedor e/ou sintetizador, que pode ser programado via API. Ao contrário do que ocorre, por exemplo, na língua inglesa, inexiste atualmente uma gama variada de recursos para o português brasileiro. O presente trabalho discute alguns esforços realizados nesse sentido, avaliando a utilização da SAPI E JSAPI, que são as APIs da Microsoft e Sun, respectivamente. Serão apresentados, outrossim, exemplos de aplicativos: uma aplicação CALL (baseada em SAPI) usando síntese em inglês e português, reconhecimento em inglês e agentes visuais; e uma proposta para agregar reconhecimento e síntese de voz ao chat IRC através de APIs Java.

Seeing Spaces: An Eye-Tracking Study Of Speech Segmentation

Speech is often a multimodal process, presented audiovisually through a talking face. One area of speech perception influenced by visual speech is speech segmentation, or the process of breaking a stream of speech into individual words. Mitchel and Weiss (2013) demonstrated that a talking face contains specific cues to word boundaries and that subjects can correctly segment a speech stream when given a silent video of a speaker. The current study expanded upon these results, using an eye tracker to identify highly attended facial features of the audiovisual display used in Mitchel and Weiss (2013). In Experiment 1, subjects were found to spend the most time watching the eyes and mouth, with a trend suggesting that the mouth was viewed more than the eyes. Although subjects displayed significant learning of word boundaries, performance was not correlated with gaze duration on any individual feature, nor was performance correlated with a behavioral measure of autistic-like traits. However, trends suggested that as autistic-like traits increased, gaze duration of the mouth increased and gaze duration of the eyes decreased, similar to significant trends seen in autistic populations (Boratston & Blakemore, 2007). In Experiment 2, the same video was modified so that a black bar covered the eyes or mouth. Both videos elicited learning of word boundaries that was equivalent to that seen in the first experiment. Again, no correlations were found between segmentation performance and SRS scores in either condition. These results, taken with those in Experiment, suggest that neither the eyes nor mouth are critical to speech segmentation and that perhaps more global head movements indicate word boundaries (see Graf, Cosatto, Strom, & Huang, 2002). Future work will elucidate the contribution of individual features relative to global head movements, as well as extend these results to additional types of speech tasks.

Visual dating of rockfall scars in Larix decidua trees

Dating past mass wasting with growth disturbances in trees is widely used in geochronology as the approach may yield dates of past process activity with up to subannual precision. Past work commonly focused on the extraction of increment cores, wedges, or stem cross sections. However, sampling has been shown to be constrained by sampling permissions, and the analysis of tree-ring samples requires considerable temporal efforts. To compensate for these shortcomings, we explore the potential of visual inspection of wound appearance for dating purposes. Based on a data set of 217 wood-penetrating wounds of known age inflicted to European larch (Larix decidua Mill.) by rockfall activity, we develop guidelines for the visual, noninvasive dating of wounds including (i) the counting of bark rings, (ii) a visual assessment of exposed wood and wound bark characteristics (such as the color and weathering status of wounds), and (iii) the relationship between wound age and tree diameter. A characterization of wounds based on photographs, randomly selected from the data set, reveals that young wounds typically can be dated with high precision, whereas dating errors gradually increase with increasing wound age. While visual dating does not reach the precision of dendrochronological dating, we clearly demonstrate that spatial patterns of and differences in rockfall activity can be reconstructed with both approaches. The introduction of visual dating approaches will facilitate fieldwork, especially in applied research, assist the conventional interpretation of tree-ring signals, and allow the reconstruction of geomorphic processes with considerably fewer temporal and financial efforts.

Neuronal Ceroid-lipofuscinosis with prominent chorea and without visual manifestations: a case report

A case of neuronal ceroid-lipofuscinosis (NCL) is reported in a 11-year-old girl, whose main symptoms were progressive dementia since the age of 4 years and choreic movements since age 10. Seizures, myoclonus and visual deterioration were absent and optic fundi were normal. A cerebral biopsy disclosed two basic types of stored substance in the cytoplasm of neurons: a) severely balloned nerve cells in cortical layers HI and V contained a non-autofluorescent material, which stained with PAS and Sudan Black B in frozen, but not in paraffin sections; ultrastructurally, these neurons showed abundant corpuscles similar to the membranous cytoplasmic bodies of Tay-Sachs disease and, in smaller amounts, also zebra bodies; b) slightly distended or non-distended neurons in all layers contained lipopigment granules, which were autofluorescent, PAS-positive and sudanophil in both frozen and paraffin sections; their ultrastructure was closely comparable to that of lipofuscin. Similar bodies were found in the swollen segments of axons and in a few astrocytes and endothelial cells. The histochemical and ultrastructural demonstration of large amounts of lipopigments allows a presumptive classification of the case as NCL. However, the presence of involuntary movements, the absence of visual disturbances and the unusual ultrastructural features place the patient into a small heterogeneous group within the NCL. A better classification of such unique instances of the disease must await elucidation of the basic enzymatic defects.

Automatic diagnosis of liver steatosis by ultrasound using autoregressive tissue characterization

Liver steatosis is mainly a textural abnormality of the hepatic parenchyma due to fat accumulation on the hepatic vesicles. Today, the assessment is subjectively performed by visual inspection. Here a classifier based on features extracted from ultrasound (US) images is described for the automatic diagnostic of this phatology. The proposed algorithm estimates the original ultrasound radio-frequency (RF) envelope signal from which the noiseless anatomic information and the textural information encoded in the speckle noise is extracted. The features characterizing the textural information are the coefficients of the first order autoregressive model that describes the speckle field. A binary Bayesian classifier was implemented and the Bayes factor was calculated. The classification has revealed an overall accuracy of 100%. The Bayes factor could be helpful in the graphical display of the quantitative results for diagnosis purposes.

Stratigraphic interpretation of Well-Log data of the Athabasca Oil Sands of Alberta Canada through Pattern recognition and Artificial Intelligence

Dissertation submitted in partial fulfillment of the requirements for the Degree of Master of Science in Geospatial Technologies.

Hand gesture recognition for human computer interaction : a comparative study of different image features

Hand gesture recognition for human computer interaction, being a natural way of human computer interaction, is an area of active research in computer vision and machine learning. This is an area with many different possible applications, giving users a simpler and more natural way to communicate with robots/systems interfaces, without the need for extra devices. So, the primary goal of gesture recognition research is to create systems, which can identify specific human gestures and use them to convey information or for device control. For that, vision-based hand gesture interfaces require fast and extremely robust hand detection, and gesture recognition in real time. In this study we try to identify hand features that, isolated, respond better in various situations in human-computer interaction. The extracted features are used to train a set of classifiers with the help of RapidMiner in order to find the best learner. A dataset with our own gesture vocabulary consisted of 10 gestures, recorded from 20 users was created for later processing. Experimental results show that the radial signature and the centroid distance are the features that when used separately obtain better results, with an accuracy of 91% and 90,1% respectively obtained with a Neural Network classifier. These to methods have also the advantage of being simple in terms of computational complexity, which make them good candidates for real-time hand gesture recognition.

Visual stimulus-dependent changes in interhemispheric EEG coherence in humans.

We analyzed the coherence of electroencephalographic (EEG) signals recorded symmetrically from the two hemispheres, while subjects (n = 9) were viewing visual stimuli. Considering the many common features of the callosal connectivity in mammals, we expected that, as in our animal studies, interhemispheric coherence (ICoh) would increase only with bilateral iso-oriented gratings located close to the vertical meridian of the visual field, or extending across it. Indeed, a single grating that extended across the vertical meridian significantly increased the EEG ICoh in normal adult subjects. These ICoh responses were obtained from occipital and parietal derivations and were restricted to the gamma frequency band. They were detectable with different EEG references and were robust across and within subjects. Other unilateral and bilateral stimuli, including identical gratings that were effective in anesthetized animals, did not affect ICoh in humans. This fact suggests the existence of regulatory influences, possibly of a top-down kind, on the pattern of callosal activation in conscious human subjects. In addition to establishing the validity of EEG coherence analysis for assaying cortico-cortical connectivity, this study extends to the human brain the finding that visual stimuli cause interhemispheric synchronization, particularly in frequencies of the gamma band. It also indicates that the synchronization is carried out by cortico-cortical connection and suggests similarities in the organization of visual callosal connections in animals and in man.

Plane-to-plane positioning from image-based visual servoing and structured light

In this paper we face the problem of positioning a camera attached to the end-effector of a robotic manipulator so that it gets parallel to a planar object. Such problem has been treated for a long time in visual servoing. Our approach is based on linking to the camera several laser pointers so that its configuration is aimed to produce a suitable set of visual features. The aim of using structured light is not only for easing the image processing and to allow low-textured objects to be treated, but also for producing a control scheme with nice properties like decoupling, stability, well conditioning and good camera trajectory

OSIRISv1.2: a named entity recognition system for sequence variants of genes in biomedical literature

Background: Single Nucleotide Polymorphisms, among other type of sequence variants, constitute key elements in genetic epidemiology and pharmacogenomics. While sequence data about genetic variation is found at databases such as dbSNP, clues about the functional and phenotypic consequences of the variations are generally found in biomedical literature. The identification of the relevant documents and the extraction of the information from them are hampered by the large size of literature databases and the lack of widely accepted standard notation for biomedical entities. Thus, automatic systems for the identification of citations of allelic variants of genes in biomedical texts are required. Results: Our group has previously reported the development of OSIRIS, a system aimed at the retrieval of literature about allelic variants of genes http://ibi.imim.es/osirisform.html. Here we describe the development of a new version of OSIRIS (OSIRISv1.2, http://ibi.imim.es/OSIRISv1.2.html webcite) which incorporates a new entity recognition module and is built on top of a local mirror of the MEDLINE collection and HgenetInfoDB: a database that collects data on human gene sequence variations. The new entity recognition module is based on a pattern-based search algorithm for the identification of variation terms in the texts and their mapping to dbSNP identifiers. The performance of OSIRISv1.2 was evaluated on a manually annotated corpus, resulting in 99% precision, 82% recall, and an F-score of 0.89. As an example, the application of the system for collecting literature citations for the allelic variants of genes related to the diseases intracranial aneurysm and breast cancer is presented. Conclusion: OSIRISv1.2 can be used to link literature references to dbSNP database entries with high accuracy, and therefore is suitable for collecting current knowledge on gene sequence variations and supporting the functional annotation of variation databases. The application of OSIRISv1.2 in combination with controlled vocabularies like MeSH provides a way to identify associations of biomedical interest, such as those that relate SNPs with diseases.

Coarse Global Visual Localization of a Mobile Robot

Localization, which is the ability of a mobile robot to estimate its position within its environment, is a key capability for autonomous operation of any mobile robot. This thesis presents a system for indoor coarse and global localization of a mobile robot based on visual information. The system is based on image matching and uses SIFT features as natural landmarks. Features extracted from training images arestored in a database for use in localization later. During localization an image of the scene is captured using the on-board camera of the robot, features are extracted from the image and the best match is searched from the database. Feature matching is done using the k-d tree algorithm. Experimental results showed that localization accuracy increases with the number of training features used in the training database, while, on the other hand, increasing number of features tended to have a negative impact on the computational time. For some parts of the environment the error rate was relatively high due to a strong correlation of features taken from those places across the environment.

ÄÄNEEN PERUSTUVA KÄYTTÄJÄN TODENTAMINEN PUHELINVERKON LISÄARVOPALVELUISSA

Tässä diplomityössä perehdytään puhujantunnistukseen ja sen käyttökelpoisuuteen käyttäjän henkilöllisyyden todentamisessa osana puhelinverkon lisäarvopalveluja. Puhelimitse ohjattavat palvelut ovat yleensä perustuneet puhelimen näppäimillä lähetettäviin äänitaajuusvalintoihin. Käyttäjän henkilöllisyydestä on voitu varmistua esimerkiksi käyttäjätunnuksen ja salaisen tunnusluvun perusteella. Tulevaisuudessa palvelut voivat perustua puheentunnistukseen, jolloin myös käyttäjän todentaminen äänen perusteella vaikuttaa järkevältä. Työssä esitellään aluksi erilaisia biometrisiä tunnistamismenetelmiä. Työssä perehdytään tarkemmin äänen perusteella tapahtuvaan puhujan todentamiseen. Työn käytännön osuudessa toteutettiin puhelinverkon palveluihin soveltuva puhujantodennussovelluksen prototyyppi. Työn tarkoituksena oli selvittää teknologian käyttömahdollisuuksia sekä kerätä kokemusta puhujantodennuspalvelun toteuttamisesta tulevaisuutta silmällä pitäen. Prototyypin toteutuksessa ohjelmointikielenä käytettiin Javaa.