Análisis de métodos de parametrización y clasificación para la simulación de un sistema de evaluación perceptual del grado de afección en voces patológicas

Los procedimientos de evaluación de la calidad de la voz basados en la valoración subjetiva a través de la percepción acústica por parte de un experto están bastante extendidos. Entre ellos,el protocolo GRBAS es el más comúnmente utilizado en la rutina clínica. Sin embargo existen varios problemas derivados de este tipo de estimaciones, el primero de los cuales es que se precisa de profesionales debidamente entrenados para su realización. Otro inconveniente reside en el hecho de que,al tratarse de una valoración subjetiva, múltiples circunstancias significativas influyen en la decisión final del evaluador, existiendo en muchos casos una variabilidad inter-evaluador e intra-evaluador en los juicios. Por estas razones se hace necesario el uso de parámetros objetivos que permitan realizar una valoración de la calidad de la voz y la detección de diversas patologías. Este trabajo tiene como objetivo comparar la efectividad de diversas técnicas de cálculo de parámetros representativos de la voz para su uso en la clasificación automática de escalas perceptuales. Algunos parámetros analizados serán los coeficientes Mel-Frequency Cepstral Coefficients(MFCC),las medidas de complejidad y las de ruido.Así mismo se introducirá un nuevo conjunto de características extraídas del Espectro de Modulación (EM) denominadas Centroides del Espectro de Modulación (CEM).En concreto se analizará el proceso de detección automática de dos de los cinco rasgos que componen la escala GRBAS: G y R. A lo largo de este documento se muestra cómo las características CEM proporcionan resultados similares a los de otras técnicas anteriormente utilizadas y propician en algún caso un incremento en la efectividad de la clasificación cuando son combinados con otros parámetros.

Codificador de voz MIDI

El presente proyecto tiene el objetivo de facilitar la composición de canciones mediante la creación de las distintas pistas MIDI que la forman. Se implementan dos controladores. El primero, con objeto de transcribir la parte melódica, convierte la voz cantada o tarareada a eventos MIDI. Para ello, y tras el estudio de las distintas técnicas del cálculo del tono (pitch), se implementará una técnica con ciertas variaciones basada en la autocorrelación. También se profundiza en el segmentado de eventos, en particular, una técnica basada en el análisis de la derivada de la envolvente. El segundo, dedicado a la base rítmica de la canción, permite la creación de la percusión mediante el golpe rítmico de objetos que disponga el usuario, que serán asignados a los distintos elementos de percusión elegidos. Los resultados de la grabación de estos impactos serán señales de corta duración, no lineales y no armónicas, dificultando su discriminación. La herramienta elegida para la clasificación de los distintos patrones serán las redes neuronales artificiales (RNA). Se realizara un estudio de la metodología de diseño de redes neuronales especifico para este tipo de señales, evaluando la importancia de las variables de diseño como son el número de capas ocultas y neuronas en cada una de ellas, algoritmo de entrenamiento y funciones de activación. El estudio concluirá con la implementación de dos redes de diferente naturaleza. Una red de Elman, cuyas propiedades de memoria permiten la clasificación de patrones temporales, procesará las cualidades temporales analizando el ataque de su forma de onda. Una red de propagación hacia adelante feed-forward, que necesitará de robustas características espectrales y temporales para su clasificación. Se proponen 26 descriptores como los derivados de los momentos del espectro: centroide, curtosis y simetría, los coeficientes cepstrales de la escala de Mel (MFCCs), y algunos temporales como son la tasa de cruces por cero y el centroide de la envolvente temporal. Las capacidades de discriminación inter e intra clase de estas características serán evaluadas mediante un algoritmo de selección, habiéndose elegido RELIEF, un método basado en el algoritmo de los k vecinos mas próximos (KNN). Ambos controladores tendrán función de trabajar en tiempo real y offline, permitiendo tanto la composición de canciones, como su utilización como un instrumento más junto con mas músicos. ABSTRACT. The aim of this project is to make song composition easier by creating each MIDI track that builds it. Two controllers are implemented. In order to transcribe the melody, the first controler converts singing voice or humming into MIDI files. To do this a technique based on autocorrelation is implemented after having studied different pitch detection methods. Event segmentation has also been dealt with, to be more precise a technique based on the analysis of the signal's envelope and it's derivative have been used. The second one, can be used to make the song's rhythm . It allows the user, to create percussive patterns by hitting different objects of his environment. These recordings results in short duration, non-linear and non-harmonic signals. Which makes the classification process more complicated in the traditional way. The tools to used are the artificial neural networks (ANN). We will study the neural network design to deal with this kind of signals. The goal is to get a design methodology, paying attention to the variables involved, as the number of hidden layers and neurons in each, transfer functions and training algorithm. The study will end implementing two neural networks with different nature. Elman network, which has memory properties, is capable to recognize sequences of data and analyse the impact's waveform, precisely, the attack portion. A feed-forward network, needs strong spectral and temporal features extracted from the hit. Some descriptors are proposed as the derivates from the spectrum moment as centroid, kurtosis and skewness, the Mel-frequency cepstral coefficients, and some temporal features as the zero crossing rate (zcr) and the temporal envelope's centroid. Intra and inter class discrimination abilities of those descriptors will be weighted using the selection algorithm RELIEF, a Knn (K-nearest neighbor) based algorithm. Both MIDI controllers can be used to compose, or play with other musicians as it works on real-time and offline.

GMM-based classifiers for the automatic detection of obstructive sleep apnea

The aim of automatic pathological voice detection systems is to serve as tools, to medical specialists, for a more objective, less invasive and improved diagnosis of diseases. In this respect, the gold standard for those system include the usage of a optimized representation of the spectral envelope, either based on cepstral coefficients from the mel-scaled Fourier spectral envelope (Mel-Frequency Cepstral Coefficients) or from an all-pole estimation (Linear Prediction Coding Cepstral Coefficients) forcharacterization, and Gaussian Mixture Models for posterior classification. However, the study of recently proposed GMM-based classifiers as well as Nuisance mitigation techniques, such as those employed in speaker recognition, has not been widely considered inpathology detection labours. The present work aims at testing whether or not the employment of such speaker recognition tools might contribute to improve system performance in pathology detection systems, specifically in the automatic detection of Obstructive Sleep Apnea. The testing procedure employs an Obstructive Sleep Apnea database, in conjunction with GMM-based classifiers looking for a better performance. The results show that an improved performance might be obtained by using such approach.

Relevance of the glottal pulse and the vocal tract in gender detection

Gender detection is a very important objective to improve efficiency in tasks as speech or speaker recognition, among others. Traditionally gender detection has been focused on fundamental frequency (f0) and cepstral features derived from voiced segments of speech. The methodology presented here consists in obtaining uncorrelated glottal and vocal tract components which are parameterized as mel-frequency coefficients. K-fold and cross-validation using QDA and GMM classifiers showed that better detection rates are reached when glottal source and vocal tract parameters are used in a gender-balanced database of running speech from 340 speakers.

Fetal origins of the TEL-AML1 fusion gene in identical twins with leukemia

The TEL (ETV6)−AML1 (CBFA2) gene fusion is the most common reciprocal chromosomal rearrangement in childhood cancer occurring in ≈25% of the most predominant subtype of leukemia— common acute lymphoblastic leukemia. The TEL-AML1 genomic sequence has been characterized in a pair of monozygotic twins diagnosed at ages 3 years, 6 months and 4 years, 10 months with common acute lymphoblastic leukemia. The twin leukemic DNA shared the same unique (or clonotypic) but nonconstitutive TEL-AML1 fusion sequence. The most plausible explanation for this finding is a single cell origin of the TEL-AML fusion in one fetus in utero, probably as a leukemia-initiating mutation, followed by intraplacental metastasis of clonal progeny to the other twin. Clonal identity is further supported by the finding that the leukemic cells in the two twins shared an identical rearranged IGH allele. These data have implications for the etiology and natural history of childhood leukemia.

Time-dependent vascular regression and permeability changes in established human tumor xenografts induced by an anti-vascular endothelial growth factor/vascular permeability factor antibody

The hyperpermeability of tumor vessels to macromolecules, compared with normal vessels, is presumably due to vascular endothelial growth factor/vascular permeability factor (VEGF/VPF) released by neoplastic and/or host cells. In addition, VEGF/VPF is a potent angiogenic factor. Removal of this growth factor may reduce the permeability and inhibit tumor angiogenesis. To test these hypotheses, we transplanted a human glioblastoma (U87), a human colon adenocarcinoma (LS174T), and a human melanoma (P-MEL) into two locations in immunodeficient mice: the cranial window and the dorsal skinfold chamber. The mice bearing vascularized tumors were treated with a bolus (0.2 ml) of either a neutralizing antibody (A4.6.1) (492 μg/ml) against VEGF/VPF or PBS (control). We found that tumor vascular permeability to albumin in antibody-treated groups was lower than in the matched controls and that the effect of the antibody was time-dependent and influenced by the mode of injection. Tumor vascular permeability did not respond to i.p. injection of the antibody until 4 days posttreatment. However, the permeability was reduced within 6 h after i.v. injection of the same amount of antibody. In addition to the reduction in vascular permeability, the tumor vessels became smaller in diameter and less tortuous after antibody injections and eventually disappeared from the surface after four consecutive treatments in U87 tumors. These results demonstrate that tumor vascular permeability can be reduced by neutralization of endogenous VEGF/VPF and suggest that angiogenesis and the maintenance of integrity of tumor vessels require the presence of VEGF/VPF in the tissue microenvironment. The latter finding reveals a new mechanism of tumor vessel regression—i.e., blocking the interactions between VEGF/VPF and endothelial cells or inhibiting VEGF/VPF synthesis in solid tumors causes dramatic reduction in vessel diameter, which may block the passage of blood elements and thus lead to vascular regression.