No description available.

To answer the question who spoke when, speaker diarization (SD) is a critical step for many speech applications in practice. The task of our project is building a MFCC-vector based speaker diarization system on top of a speaker verification system (SV), which is an existing Call-centers application to check the customer’s identity from a phone call. Our speaker diarization system uses 13-Dimensional MFCCs as Features, performs Voice Active Detection (VAD), segmentation, Linear Clustering and the Hierarchical Clustering based on GMM and the BIC score. By applying it, we decrease the Equal Error Rate (EER) of the SV from 18.1% in the baseline experiment to 3.26% on the general call-center conversations. To better analyze and evaluate the system, we also simulated a set of call-center data based on the public audio databases ICSI corpus. / För att svara på frågan vem som talade när är högtalardarisering (SD) ett kritiskt steg för många talapplikationer i praktiken. Uppdraget med vårt projekt är att bygga ett MFCC-vektorbaserat högtalar-diariseringssystem ovanpå ett högtalarverifieringssystem (SV), som är ett befintligt Call-center-program för att kontrollera kundens identitet från ett telefonsamtal. Vårt högtalarsystem använder 13-dimensionella MFCC: er som funktioner, utför Voice Active Detection (VAD), segmentering, linjär gruppering och hierarkisk gruppering baserat på GMM och BIC-poäng. Genom att tillämpa den minskar vi EER (Equal Error Rate) från 18,1 % i baslinjeexperimentet till 3,26 % för de allmänna samtalscentret. För att bättre analysera och utvärdera systemet simulerade vi också en uppsättning callcenter-data baserat på de offentliga ljuddatabaserna ICSI corpus.

No description available.

[ES] El procesado de señal sobre grafos es un campo emergente de técnicas que combinan conceptos de dos áreas muy consolidadas: el procesado de señal y la teoría de grafos. Desde la perspectiva del procesado de señal puede obtenerse una definición de la señal mucho más general asignando cada valor de la misma a un vértice de un grafo. Las señales convencionales pueden considerarse casos particulares en los que los valores de cada muestra se asignan a una cuadrícula uniforme (temporal o espacial). Desde la perspectiva de la teoría de grafos, se pueden definir nuevas transformaciones del grafo de forma que se extiendan los conceptos clásicos del procesado de la señal como el filtrado, la predicción y el análisis espectral. Además, el procesado de señales sobre grafos está encontrando nuevas aplicaciones en las áreas de detección y clasificación debido a su flexibilidad para modelar dependencias generales entre variables. En esta tesis se realizan nuevas contribuciones al procesado de señales sobre grafos. En primer lugar, se plantea el problema de estimación de la matriz Laplaciana asociada a un grafo, que determina la relación entre nodos. Los métodos convencionales se basan en la matriz de precisión, donde se asume implícitamente Gaussianidad. En esta tesis se proponen nuevos métodos para estimar la matriz Laplaciana a partir de las correlaciones parciales asumiendo respectivamente dos modelos no Gaussianos diferentes en el espacio de las observaciones: mezclas gaussianas y análisis de componentes independientes. Los métodos propuestos han sido probados con datos simulados y con datos reales en algunas aplicaciones biomédicas seleccionadas. Se demuestra que pueden obtenerse mejores estimaciones de la matriz Laplaciana con los nuevos métodos propuestos en los casos en que la Gaussianidad no es una suposición correcta. También se ha considerado la generación de señales sintéticas en escenarios donde la escasez de señales reales puede ser un problema. Los modelos sobre grafos permiten modelos de dependencia por pares más generales entre muestras de señal. Así, se propone un nuevo método basado en la Transformada de Fourier Compleja sobre Grafos y en el concepto de subrogación. Se ha aplicado en el desafiante problema del reconocimiento de gestos con las manos. Se ha demostrado que la extensión del conjunto de entrenamiento original con réplicas sustitutas generadas con los métodos sobre grafos, mejora significativamente la precisión del clasificador de gestos con las manos. / [CAT] El processament de senyal sobre grafs és un camp emergent de tècniques que combinen conceptes de dues àrees molt consolidades: el processament de senyal i la teoria de grafs. Des de la perspectiva del processament de senyal pot obtindre's una definició del senyal molt més general assignant cada valor de la mateixa a un vèrtex d'un graf. Els senyals convencionals poden considerar-se casos particulars en els quals els valors de la mostra s'assignen a una quadrícula uniforme (temporal o espacial). Des de la perspectiva de la teoria de grafs, es poden definir noves transformacions del graf de manera que s'estenguen els conceptes clàssics del processament del senyal com el filtrat, la predicció i l'anàlisi espectral. A més, el processament de senyals sobre grafs està trobant noves aplicacions en les àrees de detecció i classificació a causa de la seua flexibilitat per a modelar dependències generals entre variables. En aquesta tesi es donen noves contribucions al processament de senyals sobre grafs. En primer lloc, es planteja el problema d'estimació de la matriu Laplaciana associada a un graf, que determina la relació entre nodes. Els mètodes convencionals es basen en la matriu de precisió, on s'assumeix implícitament la gaussianitat. En aquesta tesi es proposen nous mètodes per a estimar la matriu Laplaciana a partir de les correlacions parcials assumint respectivament dos models no gaussians diferents en l'espai d'observació: mescles gaussianes i anàlisis de components independents. Els mètodes proposats han sigut provats amb dades simulades i amb dades reals en algunes aplicacions biomèdiques seleccionades. Es demostra que poden obtindre's millors estimacions de la matriu Laplaciana amb els nous mètodes proposats en els casos en què la gaussianitat no és una suposició correcta. També s'ha considerat el problema de generar senyals sintètics en escenaris on l'escassetat de senyals reals pot ser un problema. Els models sobre grafs permeten models de dependència per parells més generals entre mostres de senyal. Així, es proposa un nou mètode basat en la Transformada de Fourier Complexa sobre Grafs i en el concepte de subrogació. S'ha aplicat en el desafiador problema del reconeixement de gestos amb les mans. S'ha demostrat que l'extensió del conjunt d'entrenament original amb rèpliques substitutes generades amb mètodes sobre grafs, millora significativament la precisió del classificador de gestos amb les mans. / [EN] Graph signal processing appears as an emerging field of techniques that combine concepts from two highly consolidated areas: signal processing and graph theory. From the perspective of signal processing, it is possible to achieve a more general signal definition by assigning each value of the signal to a vertex of a graph. Conventional signals can be considered particular cases where the sample values are assigned to a uniform (temporal or spatial) grid. From the perspective of graph theory, new transformations of the graph can be defined in such a way that they extend the classical concepts of signal processing such as filtering, prediction and spectral analysis. Furthermore, graph signal processing is finding new applications in detection and classification areas due to its flexibility to model general dependencies between variables. In this thesis, new contributions are given to graph signal processing. Firstly, it is considered the problem of estimating the Laplacian matrix associated with a graph, which determines the relationship between nodes. Conventional methods are based on the precision matrix, where Gaussianity is implicitly assumed. In this thesis, new methods to estimate the Laplacian matrix from the partial correlations are proposed respectively assuming two different non-Gaussian models in the observation space: Gaussian Mixtures and Independent Component Analysis. The proposed methods have been tested with simulated data and with real data in some selected biomedical applications. It is demonstrate that better estimates of the Laplacian matrix can be obtained with the new proposed methods in cases where Gaussianity is not a correct assumption. The problem of generating synthetic signal in scenarios where real signals scarcity can be an issue has also been considered. Graph models allow more general pairwise dependence models between signal samples. Thus a new method based on the Complex Graph Fourier Transform and on the concept of subrogation is proposed. It has been applied in the challenging problem of hand gesture recognition. It has been demonstrated that extending the original training set with graph surrogate replicas, significantly improves the accuracy of the hand gesture classifier. / Belda Valls, J. (2022). Nuevas contribuciones a la teoría y aplicación del procesado de señal sobre grafos [Tesis doctoral]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/191333

<p dir="ltr">Advancements in machine learning, especially deep learning, have facilitated the creation of models capable of performing tasks previously thought impossible. This progress has opened new possibilities across diverse fields such as medical imaging and remote sensing. However, the performance of these models relies heavily on the availability of extensive labeled datasets.<br>Collecting large amounts of labeled data poses a significant financial burden, particularly in specialized fields like medical imaging and remote sensing, where annotation requires expert knowledge. To address this challenge, various methods have been developed to mitigate the necessity for labeled data or leverage information contained in unlabeled data. These encompass include self-supervised learning, few-shot learning, and semi-supervised learning. This dissertation centers on the application of semi-supervised learning in segmentation tasks.<br><br>We focus on panoptic segmentation, a task that combines semantic segmentation (assigning a class to each pixel) and instance segmentation (grouping pixels into different object instances). We choose two segmentation tasks in different domains: nerve segmentation in microscopic imaging and hyperspectral segmentation in satellite images from Mars.<br>Our study reveals that, while direct application of methods developed for natural images may yield low performance, targeted modifications or the development of robust models can provide satisfactory results, thereby unlocking new applications like machine-assisted annotation of new data.<br><br>This dissertation begins with a challenging panoptic segmentation problem in microscopic imaging, systematically exploring model architectures to improve generalization. Subsequently, it investigates how semi-supervised learning may mitigate the need for annotated data. It then moves to hyperspectral imaging, introducing a Hierarchical Bayesian model (HBM) to robustly classify single pixels. Key contributions of include developing a state-of-the-art U-Net model for nerve segmentation, improving the model's ability to segment different cellular structures, evaluating semi-supervised learning methods in the same setting, and proposing HBM for hyperspectral segmentation. <br>The dissertation also provides a dataset of labeled CRISM pixels and mineral detections, and a software toolbox implementing the full HBM pipeline, to facilitate the development of new models.</p>

Perfect simulations are methods, which convert suitable Markov chain Monte Carlo (MCMC) algorithms into algorithms which return exact draws from the target distribution, instead of approximations based on long-time convergence to equilibrium. In recent years a lot of various perfect simulation algorithms were developed. This work provides a unified exposition of some perfect simulation algorithms with applications to spatial point processes, especially to the Strauss process and area-interaction process. Described algorithms and their properties are compared theoretically and also by a simulation study.

Cette thèse s'inscrit dans le contexte de l'analyse des signaux biomédicaux multicapteurs par des méthodes stochastiques. Les signaux auxquels nous nous intéressons présentent un caractère oscillant transitoire bien représenté par les décompositions dans le plan temps-fréquence c'est pourquoi nous avons choisi de considérer non plus les décours temporels de ces signaux mais les coefficients issus de la décomposition de ces derniers dans le plan temps-fréquence. Dans une première partie, nous décomposons les signaux multicapteurs sur une base de cosinus locaux (appelée base MDCT) et nous modélisons les coefficients à l'aide d'un modèle à états latents. Les coefficients sont considérés comme les réalisations de processus aléatoires gaussiens multivariés dont la distribution est gouvernée par une chaîne de Markov cachée. Nous présentons les algorithmes classiques liés à l'utilisation des modèles de Markov caché et nous proposons une extension dans le cas où les matrices de covariance sont factorisées sous forme d'un produit de Kronecker. Cette modélisation permet de diminuer la complexité des méthodes de calcul numérique utilisées tout en stabilisant les algorithmes associés. Nous appliquons ces modèles à des données électroencéphalographiques et nous montrons que les matrices de covariance représentant les corrélations entre les capteurs et les fréquences apportent des informations pertinentes sur les signaux analysés. Ceci est notamment illustré par un cas d'étude sur la caractérisation de la désynchronisation des ondes alpha dans le contexte de la sclérose en plaques. / This thesis adresses the problem of multichannel biomedical signals analysis using stochastic methods. EEG signals exhibit specific features that are both time and frequency localized, which motivates the use of time-frequency signal representations. In this document the (time-frequency labelled) coefficients are modelled as multivariate random variables. In the first part of this work, multichannel signals are expanded using a local cosine basis (called MDCT basis). The approach we propose models the distribution of time-frequency coefficients (here MDCT coefficients) in terms of latent variables by the use of a hidden Markov model. In the framework of application to EEG signals, the latent variables describe some hidden mental state of the subject. The latter control the covariance matrices of Gaussian vectors of fixed-time vectors of multi-channel, multi-frequency, MDCT coefficients. After presenting classical algorithms to estimate the parameters, we define a new model in which the (space-frequency) covariance matrices are expanded as tensor products (also named Kronecker products) of frequency and channels matrices. Inference for the proposed model is developped and yields estimates for the model parameters, together with maximum likelihood estimates for the sequences of latent variables. The model is applied to electroencephalogram data, and it is shown that variance-covariance matrices labelled by sensor and frequency indices can yield relevant informations on the analyzed signals. This is illustrated with a case study, namely the detection of alpha waves in rest EEG for multiple sclerosis patients and control subjects.

Speech as a medium for communication has always been important in its ability to convey our ideas, personality and emotions. It is therefore not strange that Quality of Experience (QoE) becomes central to any business relying on voice communication. Using Unified Communication (UC) systems, users can communicate with each other in several ways using many different devices, making QoE an important aspect for such systems. For this thesis, automatic methods for assessing speech quality of the voice calls in Briteback’s UC application is studied, including a comparison of the researched methods. Three methods all using a Gaussian Mixture Model (GMM) as a regressor, paired with extraction of Human Factor Cepstral Coefficients (HFCC), Gammatone Frequency Cepstral Coefficients (GFCC) and Modified Mel Frequency Cepstrum Coefficients (MMFCC) features respectively is studied. The method based on HFCC feature extraction shows better performance in general compared to the two other methods, but all methods show comparatively low performance compared to literature. This most likely stems from implementation errors, showing the difference between theory and practice in the literature, together with the lack of reference implementations. Further work with practical aspects in mind, such as reference implementations or verification tools can make the field more popular and increase its use in the real world.

A urbanização e a expansão das fronteiras agrícolas na região Neotropical estão entre as principais forças causadoras da degradação ambiental em hábitats abertos naturais. Inferências e estimativas de abundância são críticas para quantificação de dinâmicas populacionais e impactos de mudanças ambientais. Contudo, a detecção imperfeita e outros fenômenos que causam inflação de zeros podem induzir erros de estimativas e dificultar a identificação de padrões ecológicos. Examinamos como a consideração desses fenômenos em dados de contagens de indivíduos não marcados pode informar na escolha do método apropriado para estimativas populacionais. Revisamos métodos estabelecidos (modelos lineares generalizados [GLMs] e amostragem de distância [distance sampling]) e emergentes que usam modelos hierárquicos baseados em misturas (N-mixture; modelo de Royle-Nichols [RN], e N-mixture básico, zero inflacionado, espacialmente explicito, visita única, e multiespécies) para estimar a abundância de populações não marcadas. Como estudo de caso, aplicamos o método N-mixture baseado em visitas únicas para modelar dados de contagens de aves de rapina em estradas e investigar como transformações de habitat no Cerrado e Pantanal do Mato Grosso do Sul afetaram as populações de 12 espécies em uma escala regional (>300.000 km2). Os métodos diferem nos pré-requisitos de desenho amostral, e a sua adequabilidade depender da espécie em questão, da escala e objetivos do estudo, e considerações financeiras e logísticas, que devem ser avaliados para que verbas, tempo e esforço sejam utilizados com eficiência. No estudo de caso, a detecção de todas as espécies foi influenciada pela horário de amostragem, com efeitos congruentes com expectativas baseadas no comportamentos de forregeamento e de voo. A vegetação fechada e carcaças também influenciaram a detecção de algumas espécies. A abundância da maioria das espécies foi negativamente influenciada pela conversão de habitats naturais para antrópicos, particularmente pastagens e plantações de soja e cana-de-açúcar, até mesmo para espécies generalistas consideradas como indicadores ruins da qualidade de hábitats. A proteção dos hábitats naturais remanescentes é essencial para prevenir um declínio ainda maior das populações de aves de rapina na área de estudo, especialmente no domínio do Cerrado / Urbanization and the expansion of agricultural frontiers are among the main forces driving the degradation of natural habitats in Neotropical open habitats. Inference and estimates of abundance are critical for quantifying population dynamics and the impacts of environmental change. Yet imperfect detection and other phenomena that cause zero inflation can induce estimation error and obscure ecological patterns. We examine how detection error and zero-inflation in count data of unmarked individuals inform the choice of analytical method for estimating population size. We review established (GLMs and distance sampling) and emerging methods that use N-mixture models (Royle-Nichols model, and basic, zero-inflated, temporary emigration, beta-binomial, generalized open-population, spatially explicit, single-visit and multispecies) to estimate abundance of unmarked populations. As a case study, we employed a single visit N-mixture approach to model roadside raptor count data and investigate how land-use transformations in the Cerrado and Pantanal domains in Brazil have affected the populations of 12 species on a regional scale (>300,000 km2). Methods differ in sampling design requirements, and their suitability will depend on the study species, scale and objectives of the study, and financial and logistical considerations, which should be evaluated to use funds, time and effort efficiently. In the case study, detection of all species was influenced by time of day, with effects that follow expectations based on foraging and flying behavior. Closed vegetation on and carcasses found during surveys also influenced detection of some species. Abundance of most species was negatively influenced by conversion of natural Cerrado and Pantanal habitats to anthropogenic uses, particularly pastures, soybean and sugar cane plantations, even for generalist species usually considered poor habitat-quality indicators. Protection of the remaining natural habitats is essential to prevent further decline of raptor populations in the study area, especially in the Cerrado domain

Les différences entre conditions d'apprentissage et conditions de test peuvent considérablement dégrader la qualité des transcriptions produites par un système de reconnaissance automatique de la parole (RAP). L'adaptation est un moyen efficace pour réduire l'inadéquation entre les modèles du système et les données liées à un locuteur ou un canal acoustique particulier. Il existe deux types dominants de modèles acoustiques utilisés en RAP : les modèles de mélanges gaussiens (GMM) et les réseaux de neurones profonds (DNN). L'approche par modèles de Markov cachés (HMM) combinés à des GMM (GMM-HMM) a été l'une des techniques les plus utilisées dans les systèmes de RAP pendant de nombreuses décennies. Plusieurs techniques d'adaptation ont été développées pour ce type de modèles. Les modèles acoustiques combinant HMM et DNN (DNN-HMM) ont récemment permis de grandes avancées et surpassé les modèles GMM-HMM pour diverses tâches de RAP, mais l'adaptation au locuteur reste très difficile pour les modèles DNN-HMM. L'objectif principal de cette thèse est de développer une méthode de transfert efficace des algorithmes d'adaptation des modèles GMM aux modèles DNN. Une nouvelle approche pour l'adaptation au locuteur des modèles acoustiques de type DNN est proposée et étudiée : elle s'appuie sur l'utilisation de fonctions dérivées de GMM comme entrée d'un DNN. La technique proposée fournit un cadre général pour le transfert des algorithmes d'adaptation développés pour les GMM à l'adaptation des DNN. Elle est étudiée pour différents systèmes de RAP à l'état de l'art et s'avère efficace par rapport à d'autres techniques d'adaptation au locuteur, ainsi que complémentaire. / Differences between training and testing conditions may significantly degrade recognition accuracy in automatic speech recognition (ASR) systems. Adaptation is an efficient way to reduce the mismatch between models and data from a particular speaker or channel. There are two dominant types of acoustic models (AMs) used in ASR: Gaussian mixture models (GMMs) and deep neural networks (DNNs). The GMM hidden Markov model (GMM-HMM) approach has been one of the most common technique in ASR systems for many decades. Speaker adaptation is very effective for these AMs and various adaptation techniques have been developed for them. On the other hand, DNN-HMM AMs have recently achieved big advances and outperformed GMM-HMM models for various ASR tasks. However, speaker adaptation is still very challenging for these AMs. Many adaptation algorithms that work well for GMMs systems cannot be easily applied to DNNs because of the different nature of these models. The main purpose of this thesis is to develop a method for efficient transfer of adaptation algorithms from the GMM framework to DNN models. A novel approach for speaker adaptation of DNN AMs is proposed and investigated. The idea of this approach is based on using so-called GMM-derived features as input to a DNN. The proposed technique provides a general framework for transferring adaptation algorithms, developed for GMMs, to DNN adaptation. It is explored for various state-of-the-art ASR systems and is shown to be effective in comparison with other speaker adaptation techniques and complementary to them.