Development of a genre-dependent TTS system with cross-speaker speaking-style transplantation

One of the biggest challenges in speech synthesis is the production of contextually-appropriate naturally sounding synthetic voices. This means that a Text-To-Speech system must be able to analyze a text beyond the sentence limits in order to select, or even modulate, the speaking style according to a broader context. Our current architecture is based on a two-step approach: text genre identification and speaking style synthesis according to the detected discourse genre. For the final implementation, a set of four genres and their corresponding speaking styles were considered: broadcast news, live sport commentaries, interviews and political speeches. In the final TTS evaluation, the four speaking styles were transplanted to the neutral voices of other speakers not included in the training database. When the transplanted styles were compared to the neutral voices, transplantation was significantly preferred and the similarity to the target speaker was as high as 78%.

A characterisation of passive and active interactions and their influence on student's achievement using Moodle LMS Logs

During the last years, there has been much concern about learning management systems' (LMS) effectiveness when compared to traditional learning and about how to assess students' participation during the course. The tracking and monitoring capabilities of most recent LMS have made it possible to analyse every interaction in the system. The issues addressed on this study are: a) Is LMS student's interaction an indicator of academic performance?; b) Are different results in performance expected between distance and in-class LMS-supported education?; c) How can LMS interactions from logs be categorised?; d) May this categorisation detect 'learning witnesses'? To answer these questions, a set of interaction types from Moodle LMS activity record logs has been analysed during two years in online and in-class Master's degrees at the UPM. The results show partial or no evidence of influence between interaction indicators and academic performance, although the proposed categorisation may help detect learning witnesses.

Improving automatic detection of obstructive sleep apnea through nonlinear analysis of sustained speech

We present a novel approach for the detection of severe obstructive sleep apnea (OSA) based on patients' voices introducing nonlinear measures to describe sustained speech dynamics. Nonlinear features were combined with state-of-the-art speech recognition systems using statistical modeling techniques (Gaussian mixture models, GMMs) over cepstral parameterization (MFCC) for both continuous and sustained speech. Tests were performed on a database including speech records from both severe OSA and control speakers. A 10 % relative reduction in classification error was obtained for sustained speech when combining MFCC-GMM and nonlinear features, and 33 % when fusing nonlinear features with both sustained and continuous MFCC-GMM. Accuracy reached 88.5 % allowing the system to be used in OSA early detection. Tests showed that nonlinear features and MFCCs are lightly correlated on sustained speech, but uncorrelated on continuous speech. Results also suggest the existence of nonlinear effects in OSA patients' voices, which should be found in continuous speech.

Evaluation of a transplantation algorithm for expressive speech synthesis

When designing human-machine interfaces it is important to consider not only the bare bones functionality but also the ease of use and accessibility it provides. When talking about voice-based inter- faces, it has been proven that imbuing expressiveness into the synthetic voices increases signi?cantly its perceived naturalness, which in the end is very helpful when building user friendly interfaces. This paper proposes an adaptation based expressiveness transplantation system capable of copying the emotions of a source speaker into any desired target speaker with just a few minutes of read speech and without requiring the record- ing of additional expressive data. This system was evaluated through a perceptual test for 3 speakers showing up to an average of 52% emotion recognition rates relative to the natural voice recognition rates, while at the same time keeping good scores in similarity and naturality.

First steps of Roman Cement in Spain

Natural cement was patented in 1796 but it didn’t arrive in Spain until 1835. No one knows exactly where the production started in Spain, because it emerged independently at the same time in many places. Most of these outbreaks are concentrated in the north and northwest of Spain: Basque Country (Zumaya and Rezola) and Catalonia (San Celoní and San Juan de las Abadesas).Natural cement was extensively used to decorate historical buildings during the nineteenth and beginning of twentieth century in Madrid. It was the building material which realised the architects and builders dreams of mass-produced cast elements in a wide variety of styles. Its arrival replaced traditional materials that were used previously (lime, gypsum and hydraulic limes). However, its use was not extended in time, and soon it was replaced by the use of artificial Portland cements. During 20th century this building material disappeared from use. What remains is it’s memory, in thousands and thousands of “stone witnesses” in our cities. Final properties of the cement largely depend on raw materials used and its combustion temperature. However, it was characterised by an easily implementation on facade masonry, fast-setting (about 15 minutes), good resistance , an agreeable structural consistency and colour.This article aims to show first steps, evolution and decay of Natural Cement Industry in Spain and its application in Madrid.

Categorization of Anti-Patterns in Software Project Management

In the last decades, software systems have become an intrinsic element in our daily lives. Software exists in our computers, in our cars, and even in our refrigerators. Today’s world has become heavily dependent on software and yet, we still struggle to deliver quality software products, on-time and within budget. When searching for the causes of such alarming scenario, we find concurrent voices pointing to the role of the project manager. But what is project management and what makes it so challenging? Part of the answer to this question requires a deeper analysis of why software project managers have been largely ineffective. Answering this question might assist current and future software project managers in avoiding, or at least effectively mitigating, problematic scenarios that, if unresolved, will eventually lead to additional failures. This is where anti-patterns come into play and where they can be a useful tool in identifying and addressing software project management failure. Unfortunately, anti-patterns are still a fairly recent concept, and thus, available information is still scarce and loosely organized. This thesis will attempt to help remedy this scenario. The objective of this work is to help organize existing, documented software project management anti-patterns by answering our two research questions: · What are the different anti-patterns in software project management? · How can these anti-patterns be categorized?

New methodology for the integration of biometric features in speaker recognition systems applied to security environments

La cuestión principal abordada en esta tesis doctoral es la mejora de los sistemas biométricos de reconocimiento de personas a partir de la voz, proponiendo el uso de una nueva parametrización, que hemos denominado parametrización biométrica extendida dependiente de género (GDEBP en sus siglas en inglés). No se propone una ruptura completa respecto a los parámetros clásicos sino una nueva forma de utilizarlos y complementarlos. En concreto, proponemos el uso de parámetros diferentes dependiendo del género del locutor, ya que como es bien sabido, la voz masculina y femenina presentan características diferentes que deberán modelarse, por tanto, de diferente manera. Además complementamos los parámetros clásicos utilizados (MFFC extraídos de la señal de voz), con un nuevo conjunto de parámetros extraídos a partir de la deconstrucción de la señal de voz en sus componentes de fuente glótica (más relacionada con el proceso y órganos de fonación y por tanto con características físicas del locutor) y de tracto vocal (más relacionada con la articulación acústica y por tanto con el mensaje emitido). Para verificar la validez de esta propuesta se plantean diversos escenarios, utilizando diferentes bases de datos, para validar que la GDEBP permite generar una descripción más precisa de los locutores que los parámetros MFCC clásicos independientes del género. En concreto se plantean diferentes escenarios de identificación sobre texto restringido y texto independiente utilizando las bases de datos de HESPERIA y ALBAYZIN. El trabajo también se completa con la participación en dos competiciones internacionales de reconocimiento de locutor, NIST SRE (2010 y 2012) y MOBIO 2013. En el primer caso debido a la naturaleza de las bases de datos utilizadas se obtuvieron resultados cercanos al estado del arte, mientras que en el segundo de los casos el sistema presentado obtuvo la mejor tasa de reconocimiento para locutores femeninos. A pesar de que el objetivo principal de esta tesis no es el estudio de sistemas de clasificación, sí ha sido necesario analizar el rendimiento de diferentes sistemas de clasificación, para ver el rendimiento de la parametrización propuesta. En concreto, se ha abordado el uso de sistemas de reconocimiento basados en el paradigma GMM-UBM, supervectores e i-vectors. Los resultados que se presentan confirman que la utilización de características que permitan describir los locutores de manera más precisa es en cierto modo más importante que la elección del sistema de clasificación utilizado por el sistema. En este sentido la parametrización propuesta supone un paso adelante en la mejora de los sistemas de reconocimiento biométrico de personas por la voz, ya que incluso con sistemas de clasificación relativamente simples se consiguen tasas de reconocimiento realmente competitivas. ABSTRACT The main question addressed in this thesis is the improvement of automatic speaker recognition systems, by the introduction of a new front-end module that we have called Gender Dependent Extended Biometric Parameterisation (GDEBP). This front-end do not constitute a complete break with respect to classical parameterisation techniques used in speaker recognition but a new way to obtain these parameters while introducing some complementary ones. Specifically, we propose a gender-dependent parameterisation, since as it is well known male and female voices have different characteristic, and therefore the use of different parameters to model these distinguishing characteristics should provide a better characterisation of speakers. Additionally, we propose the introduction of a new set of biometric parameters extracted from the components which result from the deconstruction of the voice into its glottal source estimate (close related to the phonation process and the involved organs, and therefore the physical characteristics of the speaker) and vocal tract estimate (close related to acoustic articulation and therefore to the spoken message). These biometric parameters constitute a complement to the classical MFCC extracted from the power spectral density of speech as a whole. In order to check the validity of this proposal we establish different practical scenarios, using different databases, so we can conclude that a GDEBP generates a more accurate description of speakers than classical approaches based on gender-independent MFCC. Specifically, we propose scenarios based on text-constrain and text-independent test using HESPERIA and ALBAYZIN databases. This work is also completed with the participation in two international speaker recognition evaluations: NIST SRE (2010 and 2012) and MOBIO 2013, with diverse results. In the first case, due to the nature of the NIST databases, we obtain results closed to state-of-the-art although confirming our hypothesis, whereas in the MOBIO SRE we obtain the best simple system performance for female speakers. Although the study of classification systems is beyond the scope of this thesis, we found it necessary to analise the performance of different classification systems, in order to verify the effect of them on the propose parameterisation. In particular, we have addressed the use of speaker recognition systems based on the GMM-UBM paradigm, supervectors and i-vectors. The presented results confirm that the selection of a set of parameters that allows for a more accurate description of the speakers is as important as the selection of the classification method used by the biometric system. In this sense, the proposed parameterisation constitutes a step forward in improving speaker recognition systems, since even when using relatively simple classification systems, really competitive recognition rates are achieved.

What can we learn from voice therapy of the speaking voice for singers's voice?

Voice therapies of muscle tension dysphonia in Germany need to be increased in effectiveness by applying intensive, manualized procedures and standardized assessment protocols. The same holds true for therapies of disturbed singer's voices. According to a Cochrane review on the effectiveness of therapies of functional dysphonia neither direct nor indirect voice therapies alone but combinations of both elements are effective (Ruotsalainen et al., 2007).

Balancing management and preservation of mediterranean scattered oak woodlands (Dehesas) in human-dominated landscapes

Los sistemas agroforestales mediterráneos tienen una gran importancia ecológica y socioeconómica, y mantienen altos valores medioambientales y de diversidad biológica a la vez que producen importantes servicios ecosistémicos. Estos sistemas han sido testigos de diversos cambios rápidos y drásticos en su gestión y aprovechamiento durante el último siglo. La mayor parte de la investigación desarrollada en esta tesis doctoral ha sido llevada a cabo en las dehesas españolas. Esta tesis nos muestra: i) la evidencia de la existencia de un cambio global del estrato arbóreo y del manejo del pastoreo en el todo el área de distribución de la dehesa durante los últimos 60 años; ii) la importancia del papel que juega el arbolado disperso y el adecuado manejo del ganado en la mejora de la producción, calidad y diversidad de las comunidades herbáceas, que a su vez, un pasto herbáceo bien desarrollado es importante para la rentabilidad del sistema, evaluando estos efectos bajo distintos escenarios de clima y calidad de estación; y iii) la evidencia de la falta de regeneración en sistemas agroforestales mediterráneos bajo distintos tipos de manejo del pastoreo, y además se evalúa el crecimiento y desarrollo de las pocas plántulas existentes que serán las que aseguren la viabilidad y persistencia y de estos sistemas. El arbolado disperso de estos sistemas ha experimentado una reducción importante en su densidad arbórea y fracción de cabida cubierta durante el periodo entre 1950-1980 donde tuvieron lugar importantes transformaciones en la actividad agropecuaria. La cabaña ganadera de ovino disminuyó drásticamente en los años 70 en comparación a la de bovino que desde entonces ha aumentado progresivamente hasta la actualidad. Por otro lado, el mismo tipo de manejo del ganado doméstico (especialmente bovino) durante bastante tiempo (mínimo 30 años) provocó una reducción significativa de la densidad de las plántulas. Además la probabilidad de ocurrencia y la intensidad de daños por herbivoría fue mayor bajo pastoreo bovino (con daños más intensos y consistentes) que bajo pastoreo ovino o sin pastoreo doméstico (presencia de ciervos). También el patrón de crecimiento de las plantas jóvenes estuvo afectado por el tipo de manejo, generando plántulas achaparradas en el caso del bovino y plántulas esbeltas favoreciendo el crecimiento en altura en el caso del ovino. La presencia de un arbolado disperso generó una mayor diversidad y variación en la producción de las comunidades herbáceas según las condiciones de disponibilidad de agua. Especialmente, el ecotono como microhábitat sostuvo altos valores de diversidad herbácea. La presencia del ganado bajo pastoreo continuo de intensidad moderada a alta, especialmente el bovino, incrementó los rendimientos de producción y diversidad del estrato herbáceo. Los resultados de esta tesis nos muestran la importancia que tiene la existencia de un equilibrio entre la producción y la conservación de los sistemas agroforestales mediterráneos para obtener una producción sostenible de servicios ecosistémicos mientras se asegura la perpetuación del sistema a largo plazo. Es crucial diseñar planes de gestión incorporando objetivos de conservación que integren técnicas silvopastorales apropiadas para poder aplicar en los sistemas agroforestales mediterráneos. ABSTRACT Mediterranean scattered oak woodlands have great ecological and socio-economic importance, supporting high environmental and amenity values, and relatively rich biological diversity while producing important ecosystem services. They have been witnesses of different and fast changes developed in the last century. Most of the research developed in this dissertation has conducted within dehesas. This thesis provides: i) the global change evidence of the tree layer and grazing management experienced in the land-use range of a Mediterranean scattered oak woodland (dehesa) over the last 60 years; ii) the important role of scattered trees and adequate management grazing in the improvement of grassland yield, quality and diversity - which it is important, in turn, for the system profitability - under different climate scenarios and site quality; and iii) the lack of oak regeneration evidence under some given representative management regimes and how is the growth development of these plants to assure the viability and persistence of Mediterranean scattered oak woodlands. Tree layer experienced a significant reduction in dehesas during 1950-1980 period where the highest human impacts took place. Sheep herd decreased drastically during the 1970s and, in contrast, cattle have been increasing gradually since then. On the other hand, same livestock grazing management (especially cattle) during long time (minimum 30 years) within Mediterranean scattered oak woodlands reduced strongly the density of young oak plants and showed high probability of herbivory occurrence and intensity. Young plant growth pattern was greatly modified by livestock. Cattle grazing generated stunted plants and sheep grazing generated slender plants favoring the height growth. Microsites created by large trees modified the herbaceous yield according the water availability of the year and generated high plant diversity within herbaceous communities. Especially, ecotone microsite supported high values of herbaceous diversity. The presence of livestock species increased the herbaceous yield and maintained a more diverse community under continuous grazing at both moderate and high intensities; especially cattle. Thus, around the influence of scattered trees there is a high amount of different interactions among livestock, trees and grasslands maintaining and enhancing the quality of whole dehesa system. The results of this thesis highlight how important is balancing management and preservation of Mediterranean scattered oak woodlands to obtain the optimum ecosystem services while the system conservation is assured for a long-term. It is crucial to design management plans with conservation goals that include appropriate silvopastoral practices in Mediterranean scattered oak woodlands.

Balancing management and preservation of mediterranean scattered oak woodlands (Dehesas) in human-dominated landscapes

Los sistemas agroforestales mediterráneos tienen una gran importancia ecológica y socioeconómica, y mantienen altos valores medioambientales y de diversidad biológica a la vez que producen importantes servicios ecosistémicos. Estos sistemas han sido testigos de diversos cambios rápidos y drásticos en su gestión y aprovechamiento durante el último siglo. La mayor parte de la investigación desarrollada en esta tesis doctoral ha sido llevada a cabo en las dehesas españolas. Esta tesis nos muestra: i) la evidencia de la existencia de un cambio global del estrato arbóreo y del manejo del pastoreo en el todo el área de distribución de la dehesa durante los últimos 60 años; ii) la importancia del papel que juega el arbolado disperso y el adecuado manejo del ganado en la mejora de la producción, calidad y diversidad de las comunidades herbáceas, que a su vez, un pasto herbáceo bien desarrollado es importante para la rentabilidad del sistema, evaluando estos efectos bajo distintos escenarios de clima y calidad de estación; y iii) la evidencia de la falta de regeneración en sistemas agroforestales mediterráneos bajo distintos tipos de manejo del pastoreo, y además se evalúa el crecimiento y desarrollo de las pocas plántulas existentes que serán las que aseguren la viabilidad y persistencia y de estos sistemas. El arbolado disperso de estos sistemas ha experimentado una reducción importante en su densidad arbórea y fracción de cabida cubierta durante el periodo entre 1950-1980 donde tuvieron lugar importantes transformaciones en la actividad agropecuaria. La cabaña ganadera de ovino disminuyó drásticamente en los años 70 en comparación a la de bovino que desde entonces ha aumentado progresivamente hasta la actualidad. Por otro lado, el mismo tipo de manejo del ganado doméstico (especialmente bovino) durante bastante tiempo (mínimo 30 años) provocó una reducción significativa de la densidad de las plántulas. Además la probabilidad de ocurrencia y la intensidad de daños por herbivoría fue mayor bajo pastoreo bovino (con daños más intensos y consistentes) que bajo pastoreo ovino o sin pastoreo doméstico (presencia de ciervos). También el patrón de crecimiento de las plantas jóvenes estuvo afectado por el tipo de manejo, generando plántulas achaparradas en el caso del bovino y plántulas esbeltas favoreciendo el crecimiento en altura en el caso del ovino. La presencia de un arbolado disperso generó una mayor diversidad y variación en la producción de las comunidades herbáceas según las condiciones de disponibilidad de agua. Especialmente, el ecotono como microhábitat sostuvo altos valores de diversidad herbácea. La presencia del ganado bajo pastoreo continuo de intensidad moderada a alta, especialmente el bovino, incrementó los rendimientos de producción y diversidad del estrato herbáceo. Los resultados de esta tesis nos muestran la importancia que tiene la existencia de un equilibrio entre la producción y la conservación de los sistemas agroforestales mediterráneos para obtener una producción sostenible de servicios ecosistémicos mientras se asegura la perpetuación del sistema a largo plazo. Es crucial diseñar planes de gestión incorporando objetivos de conservación que integren técnicas silvopastorales apropiadas para poder aplicar en los sistemas agroforestales mediterráneos. ABSTRACT Mediterranean scattered oak woodlands have great ecological and socio-economic importance, supporting high environmental and amenity values, and relatively rich biological diversity while producing important ecosystem services. They have been witnesses of different and fast changes developed in the last century. Most of the research developed in this dissertation has conducted within dehesas. This thesis provides: i) the global change evidence of the tree layer and grazing management experienced in the land-use range of a Mediterranean scattered oak woodland (dehesa) over the last 60 years; ii) the important role of scattered trees and adequate management grazing in the improvement of grassland yield, quality and diversity - which it is important, in turn, for the system profitability - under different climate scenarios and site quality; and iii) the lack of oak regeneration evidence under some given representative management regimes and how is the growth development of these plants to assure the viability and persistence of Mediterranean scattered oak woodlands. Tree layer experienced a significant reduction in dehesas during 1950-1980 period where the highest human impacts took place. Sheep herd decreased drastically during the 1970s and, in contrast, cattle have been increasing gradually since then. On the other hand, same livestock grazing management (especially cattle) during long time (minimum 30 years) within Mediterranean scattered oak woodlands reduced strongly the density of young oak plants and showed high probability of herbivory occurrence and intensity. Young plant growth pattern was greatly modified by livestock. Cattle grazing generated stunted plants and sheep grazing generated slender plants favoring the height growth. Microsites created by large trees modified the herbaceous yield according the water availability of the year and generated high plant diversity within herbaceous communities. Especially, ecotone microsite supported high values of herbaceous diversity. The presence of livestock species increased the herbaceous yield and maintained a more diverse community under continuous grazing at both moderate and high intensities; especially cattle. Thus, around the influence of scattered trees there is a high amount of different interactions among livestock, trees and grasslands maintaining and enhancing the quality of whole dehesa system. The results of this thesis highlight how important is balancing management and preservation of Mediterranean scattered oak woodlands to obtain the optimum ecosystem services while the system conservation is assured for a long-term. It is crucial to design management plans with conservation goals that include appropriate silvopastoral practices in Mediterranean scattered oak woodlands.

A comparison of open-source segmentation architectures for dealing with imperfect data from the media in speech synthesis

Traditional Text-To-Speech (TTS) systems have been developed using especially-designed non-expressive scripted recordings. In order to develop a new generation of expressive TTS systems in the Simple4All project, real recordings from the media should be used for training new voices with a whole new range of speaking styles. However, for processing this more spontaneous material, the new systems must be able to deal with imperfect data (multi-speaker recordings, background and foreground music and noise), filtering out low-quality audio segments and creating mono-speaker clusters. In this paper we compare several architectures for combining speaker diarization and music and noise detection which improve the precision and overall quality of the segmentation.