No description available.

The human voice may convey meaningful information about socially and evolutionarily relevant characteristics of the vocalizer. In turn, listeners may readily evaluate personal characteristics, such as body size, on the basis of nonverbal voice features. Research investigating vocal communication of physical size in humans has focused on two salient and largely independent voice features, fundamental frequency and/or corresponding harmonics (perceived as voice pitch) and formant frequencies (resonance frequencies of the supralaryngeal vocal tract). However, the degree to which fundamental and formant frequencies reliably predict variation in body size controlling for sex and age, and their relative role in the perception or accurate estimation of body size, has to date been unclear. In the current thesis, using meta-analysis, I establish that formants reliably predict variation in men’s and women’s heights and weights. In contrast, fundamental frequency only weakly predicts men’s heights and women’s weights. These findings corroborate work on many other mammals whose vocal production, like humans, follows the source-filter model. Despite the lack of a robust physical relationship between fundamental frequency and size within sexes, I further demonstrate that listeners utilize voice pitch to accurately gauge men’s relative height. My research suggests that voice pitch indirectly facilitates accurate size assessment by providing a carrier signal (i.e., dense harmonics) for formants. This is the first evidence that pitch does not confound accurate size estimation. Finally, I demonstrate that voices with lowered pitch, but not raised pitch, are perceived as larger when projected from a low than high spatial location. These results suggest that strong cross-modal perceptual biases linking low pitch to low elevation and large size may, in some contexts, cause errors in size estimation. Taken together, this thesis provides a detailed account of human vocal communication of body size, which can play a meaningful role in sexual and social contexts. / Dissertation / Doctor of Philosophy (PhD)

This study presents detailed phonetic and acoustic analyses of the speech characteristics of two new cases of Foreign Accent Syndrome (FAS). Participants include a 48-year-old female who began speaking with an "Eastern European" accent following a traumatic brain injury, and a 45-year-old male who presented with a "British" accent following a subcortical cerebral vascular accident (CVA). Identical samples of the participants' pre- and post-morbid speech were obtained, thus affording a new level of control in the study of Foreign Accent Syndrome. The speech tasks consisted of oral readings of the Grandfather Passage and 18 real words comprised of the stop consonants /p/, /t/, /k/, /b/, /d/, /g/ combined with the peripheral vowels /i/, /a/ and /u/ and ending in a voiceless stop. Computer-based acoustic measures included: 1) voice onset time (VOT), 2) vowel durations, 3) whole word durations, 4) first, second and third formant frequencies, and 5) fundamental frequency. Formant frequencies were measured at three points in the vowel duration: a) 20%, b) 50%, and c) 80% to assess differences in vowel 'onglides' and 'offglides'. The phonetic analysis provided perceptual identification of the major phonetic features associated with the foreign quality of participant's FAS speech, while acoustic measures allowed precise quantification of these features. Results indicated evidence of backing of consonant and vowel productions for both participants. The implications for future research and clinical applications are also considered.

No description available.

No description available.

Les énergies renouvelables, l’énergie sous la forme électrique et son transport à l’aide de réseaux électriques intelligents représentent aujourd’hui des enjeux majeurs car ils ont de grands impacts environnementaux et sociétaux. Ainsi, la production, le transport et la gestion de l’énergie électrique, continuent toujours à susciter un intérêt croissant. Pour atteindre ces objectifs, plusieurs verrous technologiques doivent être levés. Au-delà des questions liées aux architectures des réseaux électriques, aux modèles, aux outils de dimensionnement, à la formalisation de caractéristiques et d’indicateurs, aux contraintes et aux critères, à la gestion et à la production décentralisée, la qualité de la puissance électrique est centrale pour la fiabilité de l’ensemble du système de distribution. Les perturbations affectent la qualité des signaux électriques et peuvent provoquer des conséquences graves sur les autres équipements connectés au réseau. Les travaux de cette thèse s’inscrivent dans ce contexte et de fait ils sont orientés vers le développement de modèles, d’indicateurs et de méthodes de traitement des signaux dédiés à la surveillance en temps-réel des performances des réseaux de distribution électrique.Cette thèse analyse la qualité de la puissance électrique, en prenant en compte plusieurs caractéristiques bien connues ainsi que leur pertinence. Les modèles des systèmes électriques et les méthodes de traitement du signal pour estimer leurs paramètres sont étudiés pour des applications en temps-réel de surveillance, de diagnostic et de contrôle sous diverses conditions. Parmi tous, la fréquence fondamentale est l’un des paramètres les plus importants pour caractériser un système de distribution électrique. En effet, sa valeur qui est censée être une constante, varie en permanence et reflète la dynamique de l’énergie électrique disponible. La fréquence peut également être affectée par certaines productions d’énergie renouvelable et peut être influencée par des mauvaises synchronisations de certains équipements. En outre, la puissance absorbée par les charges ou produite par des sources est généralement différente d’une phase à l’autre. Évidemment, la plupart des installations électriques existantes avec plusieurs phases, qu’elles soient résidentielles ou industrielles, travaillent dans des conditions déséquilibrées. Identifier les composantes symétriques de tension est dans ce cas un moyen pertinent pour quantifier le déséquilibre entre les phases d’un système électrique.De nouvelles représentations de type espace d’état et modélisant des systèmes électriques sont proposées pour estimer la fréquence fondamentale et pour identifier les composantes symétriques de tension des systèmes électriques triphasés et déséquilibrés. Le premier modèle d’espace d’état proposé considère la fréquence fondamentale comme connue ou obtenue par un autre estimateur. En contrepartie, il fournit les autres paramètres caractérisant le système électrique. Un second modèle d’état-espace est introduit. Il est original dans le sens où il ne nécessite aucune connaissance de la fréquence fondamentale. Une de ses variables d’état est directement reliée à la fréquence et permet donc de la déduire. En outre, ce nouvel espace d’état est parfaitement capable de représenter des systèmes électriques à trois phases équilibrés et non équilibrés. [...] / Renewable energy, electricity and smart grids are core subjects as they have great environmental and societal impacts. Thus, generating, transporting and managing electric energy, i.e., power, still continue to drive a growing interest. In order to properly achieve these goals, several locks must be removed. Beyond issues related to the distribution architecture, the formalization of models, sizing tools, features and indicators, constraints and criteria, decentralized generation and energy management, power quality is central for the whole grid’s reliability. Disturbances affect the power quality and can cause serious impact on other equipment connected to the grid. The work of this thesis is part of this context and focuses on the development of models, indicators, and signal processing methods for power quality monitoring in time-varying power distribution systems.This thesis analyzes the power quality including several well-known features and their relevance. Power system models and signal processing methods for estimating their parameters are investigated for the purpose of real-time monitoring, diagnostic and control tasks under various operating conditions. Among all, the fundamental frequency is one of the most important parameters of a power distribution system. Indeed, its value which is supposed to be a constant varies continuously and reflects the dynamic availability of electric power. The fundamental frequency can also be affected by renewable energy generation and by nasty synchronization of some devices. Moreover, the power absorbed by loads or produced by sources is generally different from one phase to the other one. Obviously, most of the existing residential and industrial electrical installations with several phases work under unbalanced conditions. Identifying the symmetrical components is therefore an efficient way to quantify the imbalance between the phases of a grid. New state-space representations of power systems are proposed for estimating the fundamental frequency and for identifying the voltage symmetrical components of unbalanced three-phase power systems. A first state-space representation is developed by supposing the fundamental frequency to be known or to be calculated by another estimator. In return, it provides other parameters and characteristics from the power system. Another original state-space model is introduced which does not require the fundamental frequency. Here, one state variable is a function of the frequency which can thus be deduced. Furthermore this new state-space model is perfectly are able to represent a three-phase power system in both balanced and unbalanced conditions. This not the case of lots of existing models. The advantage of the proposed state-space representation is that it gives directly access to physical parameters of the system, like the frequency and the amplitude and phase values of the voltage symmetrical components. Power systems parameters can thus be estimated in real-time by using the new state-space with an online estimation process like an Extended Kalman Filter (EKF). The digital implementation of the proposed methods presents small computational requirement, elegant recursive properties, and optimal estimations with Gaussian error statistics.The methods have been implemented and validated through various tests respecting real technical constraints and operating conditions. The methods can be integrated in active power filtering schemes or load-frequency control strategies to monitor power systems and to compensate for electrical disturbances.

Abstract The central goal of this study was to study how contrastive accent is realised phonetically in three regional varieties of Standard Spoken Finnish. Speakers from the Oulu, Turku and Tampere regions produced unaccented and contrastively accented versions of the target words. Fundamental frequencies and segment durations were measured in all the target words, and in the contrastively accented versions also the temporal distance of the F0 peak from word onset. In the unaccented words, F0 fluctuations were very small, indicating once more that in Finnish, too, mere word stress is not realised tonally. In the words with CV.CV(X) structure, the lengthening of segment durations due to stress was restricted to the initial syllable in Tampere, whereas in Oulu and Turku the lengthening extended to the second syllable. The width of the fall-rise F0 pattern realising contrastive accent was in all word structures widest in the Oulu variety, and the narrowest in the Tampere variety. In the Turku variety CV.CV(X) words, the F0 peak occurred further away from word onset than in any other words investigated. The differences in segment durations among the varieties were similar in the unaccented words and in the contrastively accented ones, with one exception: the duration of V1 in the unaccented CV.CV(X) words was the same across the varieties, but in the contrastively accented CV.CV(X) words the duration of V1 was shorter in the Turku variety than in the other varieties. The durational ratio of V1 and V2 in the Turku variety – as in the Oulu variety – was different from the durational ratio in the Tampere variety: in Turku and Oulu V2 had a longer duration than V1, whereas in Tampere V1 had a longer duration than V2. This confirms earlier observations that Turku and Oulu belong to regions in which the V2 of CV.CV(X) words is half-long (longer than V1), but Tampere does not. However, the present study shows that the relative half-long duration of the V2 of CV.CV(X) words is achieved differently in Turku and Oulu: in Turku through the short duration of V1, but in Oulu through the long duration of V2. / Tiivistelmä Tämän tutkimuksen keskeisin tavoite oli selvittää, miten kontrastiivinen aksentti toteutuu foneettisesti kolmelta eri suomen murrealueelta kotoisin olevien yleiskielisessä puheessa. Oulun, Turun ja Tampereen seuduilta kotoisin olevat koehenkilöt tuottivat tutkimuksen jokaisesta kohdesanasta sekä aksentoimattoman että kontrastiivisesti aksentoidun esiintymän. Tuotetuista kohdesanoista mitattiin perustaajuuksia, äännesegmenttien kestot sekä kontrastiivisesti aksentoituiduista sanoista F0:n huipun etäisyys sanan alusta. Aksentoimattomissa sanoissa F0:n muutokset olivat kaikissa tutkituissa suomen varieteeteissa erittäin vähäisiä, mikä taas kerran todisti, ettei suomessakaan pelkkä sanapaino toteudu tonaalisesti. Sanapainon toteutumisessa kestojen avulla oli varieteettien välisiä eroja CV.CV(X)-rakenteisissa sanoissa: Tampereen varieteetissa sanapainon toteutumisala rajoittui ensimmäiseen tavuun, mutta Turun ja Oulun varieteeteissa se ulottui myös toiseen tavuun. Kontrastiivista aksenttia toteuttavan F0:n nousu–lasku-kuvion laajuus oli kaiken rakenteisissa kohdesanoissa suurin Oulun varieteetissa, pienempi Turun varieteetissa ja kaikkein pienin Tampereen varieteetissa. Muutoin kontrastiivisen aksentin toteutumisessa F0:n avulla oli huomattavia varieteettien välisiä eroja vain CV.CV(X)-rakenteisissa sanoissa: Turun varieteetin CV.CV(X)-sanoissa F0:n huippukohta sijaitsi kauempana sanan alusta kuin kaikissa muissa tutkituissa sanoissa, eli kauempana kuin muun rakenteisissa Turun varieteetin sanoissa ja kaiken rakenteisissa Oulun ja Tampereen varieteettien sanoissa. Varieteettien väliset segmenttien kestoerot olivat samat aksentoimattomissa ja kontrastiivisesti aksentoiduissa sanoissa, lukuun ottamatta sitä, että CV.CV(X)-rakenteisten sanojen V1:n kestossa ei aksentoimattomissa sanoissa ollut varieteettien välisiä eroja, mutta kontrastiivisesti aksentoiduissa sanoissa kyseisen segmentin kesto oli lyhempi Turun varieteetissa kuin muissa varieteeteissa. Tällä tavoin Turun varieteetin kontrastiivisesti aksentoiduissa CV.CV(X)-sanoissa toteutui V1:n ja V2:n kestosuhde, joka – samoin kuin Oulun varieteetin vastaava kestosuhde – poikkeaa Tampereen varieteetin vastaavasta kestosuhteesta: Turussa ja Oulussa V2 on V1:tä pitempikestoinen, Tampereella päinvastoin V1:n kesto on V2:n kestoa suurempi. Tämä vahvistaa ne aiempien tutkimusten tulokset, että Turku ja Oulu ovat ns. puolipidennysmurteiden aluetta, mutta Tampere ei. Kuitenkin tämä tutkimus osoitti, että kontrastiivisesti aksentoitujen sanojen puolipidennys saadaan Turun varieteetissa aikaan pikemminkin lyhytkestoisen V1:n kuin pitkäkestoisen V2:n avulla, kun taas Oulun varieteetissa puolipidennys syntyy nimenomaan pitkäkestoisen V2:n avulla. Kaiken kaikkiaan suurin osa tutkimuksessa todetuista varieteettien välisistä selvistä perustaajuus- ja kestoeroista koski CV.CV(X)-rakenteisia sanoja, jotka ovatkin erikoinen suomen sanatyyppi yksimoraisen ensi tavunsa vuoksi.

Tato disertační práce je zaměřena na analýzu řečového signálu za učelem detekce nemocí ovlivňujících strukturu hlasových orgánů, obzvláště těch, které mění strukturální character hlasivek. Poskytnut je přehled současných technik. Dále jsou popsány zdroje použitých nahrávek pro zdravé a nemocné mlučí. Hlavním učelem této disertační práce je popsat vypočetní postup k odhadu parametrů modelu hlasového zdroje, které umožní následnou detekci a klasifikaci nemocí hlasivek. Poskytujeme detailní popis analýzy řečových signálů, které mohou být odvozeny z parametrických modelů hlasivek.