Towards speaking style transplantation in speech synthesis

One of the biggest challenges in speech synthesis is the production of naturally sounding synthetic voices. This means that the resulting voice must be not only of high enough quality but also that it must be able to capture the natural expressiveness imbued in human speech. This paper focus on solving the expressiveness problem by proposing a set of different techniques that could be used for extrapolating the expressiveness of proven high quality speaking style models into neutral speakers in HMM-based synthesis. As an additional advantage, the proposed techniques are based on adaptation approaches, which means that they can be used with little training data (around 15 minutes of training data are used in each style for this paper). For the final implementation, a set of 4 speaking styles were considered: news broadcasts, live sports commentary, interviews and parliamentary speech. Finally, the implementation of the 5 techniques were tested through a perceptual evaluation that proves that the deviations between neutral and speaking style average models can be learned and used to imbue expressiveness into target neutral speakers as intended.

Style Guidelines for Naming and Labeling Ontologies in the Multilingual Web

In the context of the Semantic Web, natural language descriptions associated with ontologies have proven to be of major importance not only to support ontology developers and adopters, but also to assist in tasks such as ontology mapping, information extraction, or natural language generation. In the state-of-the-art we find some attempts to provide guidelines for URI local names in English, and also some disagreement on the use of URIs for describing ontology elements. When trying to extrapolate these ideas to a multilingual scenario, some of these approaches fail to provide a valid solution. On the basis of some real experiences in the translation of ontologies from English into Spanish, we provide a preliminary set of guidelines for naming and labeling ontologies in a multilingual scenario.

Influence of driving style on fuel consumption and Emissions in diesel-powered passenger car

This paper presents the main results of a study on the influence of driving style on fuel consumption and pollutant emissions of diesel passenger car in urban traffic. Driving styles (eco, normal or aggressive) patterns were based on the “eco-driving” criteria. The methodology is based on on-board emission measurements in real urban traffic in the city of Madrid. Five diesel passenger cars, have been tested. Through a statistical analysis, a Dynamic Performance Index was defined for diesel passenger cars. Likewise, the CO, NOX and HC emissions were compared for each driving style for the tested vehicles. Eco-driving reduces by 14% fuel consumption and CO2 emissions, but aggressive driving increase consumption by 40%. Aggressive driving increases NOX emission by more than 40%. CO and HC, show different trends, but being increased in eco-driving style.

Precise set sharin analysis for java-style programs (and proofs).

Finding useful sharing information between instances in object- oriented programs has recently been the focus of much research. The applications of such static analysis are multiple: by knowing which variables definitely do not share in memory we can apply conventional compiler optimizations, find coarse-grained parallelism opportunities, or, more importantly, verify certain correctness aspects of programs even in the absence of annotations. In this paper we introduce a framework for deriving precise sharing information based on abstract interpretation for a Java-like language. Our analysis achieves precision in various ways, including supporting multivariance, which allows separating different contexts. We propose a combined Set Sharing + Nullity + Classes domain which captures which instances do not share and which ones are definitively null, and which uses the classes to refine the static information when inheritance is present. The use of a set sharing abstraction allows a more precise representation of the existing sharings and is crucial in achieving precision during interprocedural analysis. Carrying the domains in a combined way facilitates the interaction among them in the presence of multivariance in the analysis. We show through examples and experimentally that both the set sharing part of the domain as well as the combined domain provide more accurate information than previous work based on pair sharing domains, at reasonable cost.

Precise set sharing and nullity analysis for java-style program

Finding useful sharing information between instances in object- oriented programs has been recently the focus of much research. The applications of such static analysis are multiple: by knowing which variables share in memory we can apply conventional compiler optimizations, find coarse-grained parallelism opportunities, or, more importantly,erify certain correctness aspects of programs even in the absence of annotations In this paper we introduce a framework for deriving precise sharing information based on abstract interpretation for a Java-like language. Our analysis achieves precision in various ways. The analysis is multivariant, which allows separating different contexts. We propose a combined Set Sharing + Nullity + Classes domain which captures which instances share and which ones do not or are definitively null, and which uses the classes to refine the static information when inheritance is present. Carrying the domains in a combined way facilitates the interaction among the domains in the presence of mutivariance in the analysis. We show that both the set sharing part of the domain as well as the combined domain provide more accurate information than previous work based on pair sharing domains, at reasonable cost.

Towards an unsupervised speaking style voice building framework: multi-style speaker diarization

Current text-to-speech systems are developed using studio-recorded speech in a neutral style or based on acted emotions. However, the proliferation of media sharing sites would allow developing a new generation of speech-based systems which could cope with spontaneous and styled speech. This paper proposes an architecture to deal with realistic recordings and carries out some experiments on unsupervised speaker diarization. In order to maximize the speaker purity of the clusters while keeping a high speaker coverage, the paper evaluates the F-measure of a diarization module, achieving high scores (>85%) especially when the clusters are longer than 30 seconds, even for the more spontaneous and expressive styles (such as talk shows or sports).

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.

Dynamic gamma frequency feedback coupling between higher and lower order visual cortices underlies perceptual completion in humans

To perceive a coherent environment, incomplete or overlapping visual forms must be integrated into meaningful coherent percepts, a process referred to as ?Gestalt? formation or perceptual completion. Increasing evidence suggests that this process engages oscillatory neuronal activity in a distributed neuronal assembly. A separate line of evidence suggests that Gestalt formation requires top-down feedback from higher order brain regions to early visual cortex. Here we combine magnetoencephalography (MEG) and effective connectivity analysis in the frequency domain to specifically address the effective coupling between sources of oscillatory brain activity during Gestalt formation. We demonstrate that perceptual completion of two-tone ?Mooney? faces induces increased gamma frequency band power (55?71 Hz) in human early visual, fusiform and parietal cortices. Within this distributed neuronal assembly fusiform and parietal gamma oscillators are coupled by forward and backward connectivity during Mooney face perception, indicating reciprocal influences of gamma activity between these higher order visual brain regions. Critically, gamma band oscillations in early visual cortex are modulated by top-down feedback connectivity from both fusiform and parietal cortices. Thus, we provide a mechanistic account of Gestalt perception in which gamma oscillations in feature sensitive and spatial attention-relevant brain regions reciprocally drive one another and convey global stimulus aspects to local processing units at low levels of the sensory hierarchy by top-down feedback. Our data therefore support the notion of inverse hierarchical processing within the visual system underlying awareness of coherent percepts.

Depth perceptual video coding for free viewpoint video based on H.264/AVC

A novel scheme for depth sequences compression, based on a perceptual coding algorithm, is proposed. A depth sequence describes the object position in the 3D scene, and is used, in Free Viewpoint Video, for the generation of synthetic video sequences. In perceptual video coding the human visual system characteristics are exploited to improve the compression efficiency. As depth sequences are never shown, the perceptual video coding, assessed over them, is not effective. The proposed algorithm is based on a novel perceptual rate distortion optimization process, assessed over the perceptual distortion of the rendered views generated through the encoded depth sequences. The experimental results show the effectiveness of the proposed method, able to obtain a very considerable improvement of the rendered view perceptual quality.

Heavyweight extensions to the UML metamodel to describe the C3 architectural style

UML is widely accepted as the standard for representing the various software artifacts generated by a development process. For this reason, there have been attempts to use this language to represent the software architecture of systems as well. Unfortunately, these attempts have ended in the same representations (boxes and lines) already criticized by the software architecture community.In this work we propose an extension to the UML metamodel that is able to represent the syntactics and semantics of the C3 architectural style. This style is derived from C2. The modifications to define C3 are described in section 4. This proposal is innovative regarding UML extensions for software architectures, since previous proposals where based on light extensions to the UML meta-model, while we propose a heavyweight extension of the metamodel. On the other hand, this proposal is less ambitious than previous proposals, since we do not want to represent in UML any architectural style, but only one: C3.

Perceptual distortion modeling for side-by-side 3D video delivery

A frame-level distortion model based on perceptual features of the human visual system is proposed to improve the performance of unequal error protection strategies and provide better quality of experience to users in Side-by-Side 3D video delivery systems.

New experiments on speaker diarization for unsupervised speaking style voice building for speech synthesis

El uso universal de síntesis de voz en diferentes aplicaciones requeriría un desarrollo sencillo de las nuevas voces con poca intervención manual. Teniendo en cuenta la cantidad de datos multimedia disponibles en Internet y los medios de comunicación, un objetivo interesante es el desarrollo de herramientas y métodos para construir automáticamente las voces de estilo de varios de ellos. En un trabajo anterior se esbozó una metodología para la construcción de este tipo de herramientas, y se presentaron experimentos preliminares con una base de datos multiestilo. En este artículo investigamos más a fondo esta tarea y proponemos varias mejoras basadas en la selección del número apropiado de hablantes iniciales, el uso o no de filtros de reducción de ruido, el uso de la F0 y el uso de un algoritmo de detección de música. Hemos demostrado que el mejor sistema usando un algoritmo de detección de música disminuye el error de precisión 22,36% relativo para el conjunto de desarrollo y 39,64% relativo para el montaje de ensayo en comparación con el sistema base, sin degradar el factor de mérito. La precisión media para el conjunto de prueba es 90.62% desde 76.18% para los reportajes de 99,93% para los informes meteorológicos.

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%.

Modeling of GRBAS perceptual evaluation using spectral features obtained from an auditory-based filterbank.

Perceptual voice evaluation according to the GRBAS scale is modelled using a linear combination of acoustic parameters calculated after a filter-bank analysis of the recorded voice signals. Modelling results indicate that for breathiness and asthenia more than 55% of the variance of perceptual rates can be explained by such a model, with only 4 latent variables. Moreover, the greatest part of the explained variance can be attributed to only one or two latent variables similarly weighted by all 5 listeners involved in the experiment. Correlation factors between actual rates and model predictions around 0.6 are obtained.

Differential elimination by differential specialization of Sylvester style matrices

Differential resultant formulas are defined, for a system $\cP$ of $n$ ordinary Laurent differential polynomials in $n-1$ differential variables. These are determinants of coefficient matrices of an extended system of polynomials obtained from $\cP$ through derivations and multiplications by Laurent monomials. To start, through derivations, a system $\ps(\cP)$ of $L$ polynomials in $L-1$ algebraic variables is obtained, which is non sparse in the order of derivation. This enables the use of existing formulas for the computation of algebraic resultants, of the multivariate sparse algebraic polynomials in $\ps(\cP)$, to obtain polynomials in the differential elimination ideal generated by $\cP$. The formulas obtained are multiples of the sparse differential resultant defined by Li, Yuan and Gao, and provide order and degree bounds in terms of mixed volumes in the generic case.