The analysis of long-time formant distribution is relatively young but promising discipline of speaker identification. It is a method of mapping the long-term behavior of formants in speech of individual speakers. Frequently encountered problems in practice are bad acoustic quality and very short duration of analyzed recordings. This work aims to present the historical development of forensic phonetics and currently used methods. In the practical part, it deals with the usability of LTF method in forensic practice, especially in recordings containing background noise. It was shown that the noise appreciably affects extracted LTF values and unfortunately the change is not systematic. Therefore, we proposed several methods to compensate the noise in recordings, in order to be able to compare recordings with and without noise. We also investigated the minimum duration of recording, which is necessary for statistical reliability of the resulting values. This boundary is not exact and for each speaker, it is substantially individual. But it is apparent that recordings (vocalic streams) shorter than 15 s often provide incomplete information, wherefore they cannot be recommended for analysis. Keywords: LTF, long-time formant distribution, speaker identification, forensic phonetics, acoustic quality of...

The diploma thesis is a reflection on the concept of error; it defines error in regard to language norm and language usage. It deals with errors committed by high school students. In addition, it seeks to uncover at least some of the causes why and how the error occurs, and it focuses on the influence of interference. It examines whether the phenomena that do not occur in Czech, are actually more difficult to understand and then to use in practical language. In the theoretical part the work also treats the issue of fault assessment, answers the critical questions in this area and summarizes some recommendations to the motivational and constructive evaluation that leads to understanding of the problem, and thus eliminating errors. In the practical part, it introduces the text which served as a basis for the research. In addition to the translation, students were also asked questions about the style of teaching, learning, assessment and their motivation for the studying French. Conclusions drawn from both the translations and the interviews are presented at the end of work. Keywords: error, grammar, meaning, speaker, interference, evaluation, learning styles, motivation, translation, written production

This thesis deals with temporal characteristics of tone units in read speeches of professional speakers. The main goal of the research is to follow variability of the articulation rate (AR) within tone units. Thus, the domain where we investigate AR is the tone unit and we also deal with the general articulation rate. Read narrations in Czech in wildlife documentaries represented the studied material. To verify the statistical significance, we used the linear mixed effects model and the ANOVA test. The results of the measurement of the general average articulation rate within the narrations showed that the values of one speaker were significantly different from those of the other speakers. Interpersonal variability was also proved by the ANOVA tests and intrapersonal variability was also evidenced. The average AR values with regard to linear segmentation proved to be relatively uniform. Statistical verifications did not prove any statistical significance either. In our analyses of tone units, we investigated whether AR was influenced by the size of tone units in prosodic words, further by the position of the prosodic word in the tone unit, and what were the directions of AR changes inside the unit. Statistical tests were used to verify the significance of the impact of the tone unit size as well as...

My thesis deals with the status of German language standard and is based on research in the internet discussion sites (internet forums), that deal with the topic of language. The internet forums that I've chosen offer help in learning a foreign language, in our case German. The help is provided either by proofreading or by answering the specific questions. This thesis is based on Ulrich Ammon's model of four social forces and the theory of language management. Next to the instances, which according to Ammon determine the standard, I included the participants of discussion, who decide what is standard too and thereby affect those for whom isn't German the native language. The analysis will be carried out on the basis of the proofreading of model texts. With the help of the language management theory I will discuss with the proofreader in the comments their selected (or not selected) option. The aim of my work is to observe the behavior of normative authority in relation to the codified norm, whether they know and follow it. Furthermore, I am interested in the social background and language abilities of the participants, which I'll find out by using the questionnaires.

Mes travaux de thèse s'intéressent à la voix émotionnelle dans un contexte d'interaction humain-robot. Dans une interaction réaliste, nous définissons au moins quatre grands types de variabilités : l'environnement (salle, microphone); le locuteur, ses caractéristiques physiques (genre, âge, type de voix) et sa personnalité; ses états émotionnels; et enfin le type d'interaction (jeu, situation d'urgence ou de vie quotidienne). A partir de signaux audio collectés dans différentes conditions, nous avons cherché, grâce à des descripteurs acoustiques, à imbriquer la caractérisation d'un locuteur et de son état émotionnel en prenant en compte ces variabilités.Déterminer quels descripteurs sont essentiels et quels sont ceux à éviter est un défi complexe puisqu'il nécessite de travailler sur un grand nombre de variabilités et donc d'avoir à sa disposition des corpus riches et variés. Les principaux résultats portent à la fois sur la collecte et l'annotation de corpus émotionnels réalistes avec des locuteurs variés (enfants, adultes, personnes âgées), dans plusieurs environnements, et sur la robustesse de descripteurs acoustiques suivant ces quatre variabilités. Deux résultats intéressants découlent de cette analyse acoustique: la caractérisation sonore d'un corpus et l'établissement d'une liste "noire" de descripteurs très variables. Les émotions ne sont qu'une partie des indices paralinguistiques supportés par le signal audio, la personnalité et le stress dans la voix ont également été étudiés. Nous avons également mis en oeuvre un module de reconnaissance automatique des émotions et de caractérisation du locuteur qui a été testé au cours d'interactions humain-robot réalistes. Une réflexion éthique a été menée sur ces travaux. / This thesis deals with emotional voices during a human-robot interaction. In a natural interaction, we define at least, four kinds of variabilities: environment (room, microphone); speaker, its physic characteristics (gender, age, voice type) and personality; emotional states; and finally the kind of interaction (game scenario, emergency, everyday life). From audio signals collected in different conditions, we tried to find out, with acoustic features, to overlap speaker and his emotional state characterisation taking into account these variabilities.To find which features are essential and which are to avoid is hard challenge because it needs to work with a high number of variabilities and then to have riche and diverse data to our disposal. The main results are about the collection and the annotation of natural emotional corpora that have been recorded with different kinds of speakers (children, adults, elderly people) in various environments, and about how reliable are acoustic features across the four variabilities. This analysis led to two interesting aspects: the audio characterisation of a corpus and the drawing of a black list of features which vary a lot. Emotions are ust a part of paralinguistic features that are supported by the audio channel, other paralinguistic features have been studied such as personality and stress in the voice. We have also built automatic emotion recognition and speaker characterisation module that we have tested during realistic interactions. An ethic discussion have been driven on our work.

Projetar máquinas capazes identificar pessoas é um problema cuja solução encontra uma grande quantidade de aplicações. Implementações em software de sistemas baseados em medições de características físicas pessoais (biométricos), estão começando a ser produzidos em escala comercial. Nesta categoria estão os sistemas de Reconhecimento Automático de Locutor, que se usam da voz como característica identificadora. No presente momento, os métodos mais populares são baseados na extração de coeficientes mel-cepstrais (MFCCs) das locuções, seguidos da identificação do locutor através de Hidden Markov Models (HMMs), Gaussian Mixture Models (GMMs) ou quantização vetorial. Esta preferência se justifica pela qualidade dos resultados obtidos. Fazer com que estes sistemas sejam robustos, mantendo sua eficiência em ambientes ruidosos, é uma das grandes questões atuais. Igualmente relevantes são os problemas relativos à degradação de performance em aplicações envolvendo um grande número de locutores, e a possibilidade de fraude baseada em vozes gravadas. Outro ponto importante é embarcar estes sistemas como sub-sistemas de equipamentos já existentes, tornando-os capazes de funcionar de acordo com o seu operador. Este trabalho expõe os conceitos e algoritmos envolvidos na implementação de um software de Reconhecimento Automático de Locutor independente de texto. Inicialmente é tratado o processamento dos sinais de voz e a extração dos atributos essenciais deste sinal para o reconhecimento. Após isto, é descrita a forma pela qual a voz de cada locutor é modelada através de uma rede neural de arquitetura Self-Organizing Map (SOM) e o método de comparação entre as respostas dos modelos quando apresentada uma locução de um locutor desconhecido. Por fim, são apresentados o processo de construção do corpus de vozes usado para o treinamento e teste dos modelos, as arquiteturas de redes testadas e os resultados experimentais obtidos numa tarefa de identificação de locutor. / The design of machines that can identify people is a problem whose solution has a wide range of applications. Software systems, based on personal phisical attributes measurements (biometrics), are in the beginning of commercial scale production. Automatic Speaker Recognition systems fall into this cathegory, using voice as the identifying attribute. At present, the most popular methods are based on the extraction of mel-frequency cepstral coefficients (MFCCs), followed by speaker identification by Hidden Markov Models (HMMs), Gaussian Mixture Models (GMMs) or vector quantization. This preference is motivated by the quality of the results obtained by the use of these methods. Making these systems robust, able to keep themselves efficient in noisy environments, is now a major concern. Just as relevant are the problems related to performance degradation in applications with a large number of speakers involved, and the issues related to the possibility of fraud by the use of recorded voices. Another important subject is to embed these systems as sub-systems of existing devices, enabling them to work according to the operator. This work presents the relevant concepts and algorithms concerning the implementation of a text-independent Automatic Speaker Recognition software system. First, the voice signal processing and the extraction of its essential features for recognition are treated. After this, it is described the way each speaker\'s voice is represented by a Self-Organizing Map (SOM) neural network, and the comparison method of the models responses when a new utterance from an unknown speaker is presented. At last, it is described the construction of the speech corpus used for training and testing the models, the neural network architectures tested, and the experimental results obtained in a speaker identification task.

Gesicht und Stimme einer Person sind stark miteinander assoziiert und werden normalerweise als eine Einheit wahrgenommen. Trotz des natürlichen gemeinsamen Auftretens von Gesichtern und Stimmen, wurden deren Wahrnehmung in den Neurowissenschaften traditionell aus einer unisensorischen Perspektive untersucht. Das heißt, dass sich Forschung zu Gesichtswahrnehmung ausschließlich auf das visuelle System fokusierte, während Forschung zu Stimmwahrnehmung nur das auditorische System untersuchte. In dieser Arbeit schlage ich vor, dass das Gehirn an die multisensorische Beschaffenheit von Gesichtern und Stimmen adaptiert ist, und dass diese Adaption sogar dann sichtbar ist, wenn nur die Stimme einer Person gehört wird, ohne dass das Gesicht zu sehen ist. Im Besonderen, untersucht diese Arbeit wie das Gehirn zuvor gelernte Gesichts-Stimmassoziationen ausnutzt um die auditorische Analyse von Stimmen und Sprache zu optimieren. Diese Dissertation besteht aus drei empirischen Studien, welche raumzeitliche Hirnaktivität mittels funktionaler Magnetresonanztomographie (fMRT) und Magnetoenzephalographie (MEG) liefern. Alle Daten wurden gemessen, während Versuchspersonen auditive Sprachbeispiele von zuvor familiarisierten Sprechern (mit oder ohne Gesicht des Sprechers) hörten. Drei Ergebnisse zeigen, dass zuvor gelernte visuelle Sprecherinformationen zur auditorischen Analyse von Stimmen beitragen: (i) gesichtssensible Areale waren Teil des sensorischen Netzwerks, dass durch Stimmen aktiviert wurde, (ii) die auditorische Verarbeitung von Stimmen war durch die gelernte Gesichtsinformation zeitlich faszilitiert und (iii) multisensorische Interaktionen zwischen gesichtsensiblen und stimm-/sprachsensiblen Arealen waren verstärkt. Die vorliegende Arbeit stellt den traditionellen, unisensorischen Blickwinkel auf die Wahrnehmung von Stimmen und Sprache in Frage und legt nahe, dass die Wahrnehmung von Stimme und Sprache von von einem multisensorischen Verarbeitungsschema profitiert. / Face and voice of a person are strongly associated with each other and usually perceived as a single entity. Despite the natural co-occurrence of faces and voices, brain research has traditionally approached their perception from a unisensory perspective. This means that research into face perception has exclusively focused on the visual system, while research into voice perception has exclusively probed the auditory system. In this thesis, I suggest that the brain has adapted to the multisensory nature of faces and voices and that this adaptation is evident even when one input stream is missing, that is, when input is actually unisensory. Specifically, the current work investigates how the brain exploits previously learned voice-face associations to optimize the auditory processing of voices and vocal speech. Three empirical studies providing spatiotemporal brain data—via functional magnetic resonance imaging (fMRI) and magnetoencephalography (MEG)—constitute this thesis. All data were acquired while participants listened to auditory-only speech samples of previously familiarized speakers (with or without seeing the speakers’ faces). Three key findings demonstrate that previously learned visual speaker information support the auditory analysis of vocal sounds: (i) face-sensitive areas were part of the sensory network activated by voices, (ii) the auditory analysis of voices was temporally facilitated by learned facial associations and (iii) multisensory interactions between face- and voice/speech-sensitive regions were increased. The current work challenges traditional unisensory views on vocal perception and rather suggests that voice and vocal speech perception profit from a multisensory neural processing scheme.

Die Dissertation thematisiert sprecherspezifische Variabilität bei ein- und zweieiigen Zwillingen hinsichtlich Artikulation, Akustik und Perzeption. Die zentrale Fragestellung ist, ob sprecherspezifische Charakteristika auf physiologisch-biologischen Differenzen der Sprecher beruhen (BIOLOGIE), oder sich auf gelernte, umweltabhängige Unterschiede zurückführen lassen (UMWELT). Artikulatorische und akustische Daten wurden von 4 eineiigen Zwillingspaaren (EZ, 100% genetische Übereinstimmung) und 3 zweieiigen Zwillingspaaren (ZZ, 50% genetische Übereinstimmung) analysiert. Zusätzlich wurde ein Perzeptionstest zur auditiven Ähnlichkeit der Zwillinge durchgeführt. Auf einen großen Einfluss des Faktors BIOLOGIE lässt sich schließen, wenn sich EZ ähnlicher sind als ZZ. Sind sich aber ZZ genauso ähnlich wie EZ, zeigt sich die Wichtigkeit der gleichen Lernumgebung (UMWELT). Die Ergebnisse weisen auf einen großen Einfluss des Faktors UMWELT und stützen die Hypothese, dass sprachliche Ziele gelernt sind und sich am auditiven Feedback orientieren. Darüber hinaus wurden drei Faktoren gefunden, die den Einfluss der BIOLOGIE intensivieren: a) Lautklasse, b) Wortakzent und c) Koartikulation. Plosive und Sibilanten sind aufgrund des stärker ausgeprägten linguo-palatalen Kontaktes mehr durch die individuelle Physiologie beeinflusst als Vokale. Außerdem wurde ein größerer Effekt des Faktors BIOLOGIE in unbetonten als in betonten Silben gefunden. Zusätzlich stellten sich koartikulatorische Prozesse als wichtig heraus: dynamische Parameter – artikulatorische Gesten und akustische Transitionen – sind stärker durch die Physiologie beeinflusst als statische Parameter – artikulatorische Ziele und stabile akustische Regionen. Sowohl der Faktor BIOLOGIE als auch der Faktor UMWELT sind einflussreiche Größen hinsichtlich sprecherspezifischer Variabilität. Welcher der beiden Faktoren die übergeordnete Rolle übernimmt, hängt von den spezifischen Charakteristika des untersuchten Parameters ab. / This dissertation examines inter-speaker variability in monozygotic (MZ) and dizygotic (DZ) twin pairs in regard to articulation, acoustics and perception. The aim of the study is to evaluate whether speaker-specific variability reflects physiological differences between speakers (NATURE) or bases on learned variation due to social environmental influences (NURTURE). Articulatory and acoustic data was analyzed from 4 MZ twin pairs (100% identical genes) and 3 DZ twin pairs (50 % identical genes). Additionally, a perception experiment was carried out to explore the perceived auditory similarity. The effect of NATURE should have a larger impact than the effect of NURTURE, if a parameter differs more in DZ than in MZ twin pairs. If MZ and DZ twins show the same amount of inter-speaker variability, NURTURE seems to be crucial. Results point to the importance of NURTURE and shared social environment. Nevertheless, three factors were found that intensify the effect of NATURE: a) phoneme class, b) lexical stress, and c) degree of coarticulation. Somatosensory feedback plays a larger role for consonants than for vowels, and thus individual physiology was found to shape articulation more in sibilants and stops than in vowels. Additionally, a stronger impact of NATURE was found in parameters that are auditorily less salient: unstressed syllables were more similar in MZ than in DZ twins, while for stressed syllables this was not the case. Moreover, coarticulation turned out to be essential: dynamic parameters – articulatory gestures and acoustic transitions – were more influenced by physiological constraints (NATURE) than static parameters – articulatory targets and stable acoustic regions. Thus, both NATURE and NURTURE are crucial influencing factors in speaker-specific variability. However, the relative importance of the two factors is highly dependent on the specific characteristics of the investigated parameter.

Made available in DSpace on 2016-04-28T18:23:27Z (GMT). No. of bitstreams: 1 Claudia Sousa Fernandes.pdf: 623395 bytes, checksum: acb1a54152507a3f1b940aa3646c95f2 (MD5) Previous issue date: 2006-12-01 / This study aims at investigating the representations of four teachers of English from a language institute in the outskirts of São Paulo towards the English language, as well as the implications of those representations to the construction of their professional identities. The theoretical framework of this research is based on the identity concept (Moita Lopes, 2002; Rajagopalan, 1998; Signorini, 2002; among others), the representation concept (Moscovici, 2000; Freire e Lessa, 2003; Celani e Magalhães, 2002), the discussion about the native speaker (Braine, 1999; Phillipson, 1992; Widdowson, 1994; Kramsh, 1997; Rajagopalan, 2005) and the role of teacher in the major teaching and learning theories (Bruner, 1996/ 2001; Mizukami, 1986; Williams e Burden, 1997; Vygotsky 1934/ 2003; Milhollan e Forisha, 1978). The data were analyzed according to the concept of thematic content proposed by Bronckart (2003) based on the lexical choices made by the participants in order to answer the following research questions: What s the representation of the teachers in relation to the English language? What s the representation of the teachers about themselves as English language speakers? How do these representations influence the constructing of their identities? Results indicate how the analyzed representations contribute to the construction of the identities of the participants as speakers of English / Este estudo tem por objetivo investigar as representações de quatro professores de inglês de um instituto de idiomas da periferia de São Paulo sobre a língua inglesa, verificando qual ou quais as implicações dessas representações em relação à construção de suas identidades profissionais. Os pressupostos teóricos que fundamentam esta pesquisa são: o conceito de identidade (Moita Lopes, 2002; Rajagopalan, 1998; Signorini, 2002; entre outros), o conceito de representação (Moscovici, 2000; Freire e Lessa, 2003; Celani e Magalhães, 2002) a questão do falante nativo (Braine, 1999; Phillipson, 1992; Widdowson, 1994; Kramsh, 1997; Rajagopalan, 2005) e o papel do professor nas principais teorias de ensino-aprendizagem (Bruner, 1996/ 2001; Mizukami,1986; Williams e Burden, 1997; Vygotsky, 1934/ 2003; Milhollan e Forisha, 1978). Para a análise e interpretação dos dados foi utilizado o conceito de conteúdo temático de Bronckart (2003), com base nas escolhas lexicais dos participantes focais desta pesquisa apresentados no início deste resumo com o objetivo de responder a três perguntas de pesquisa: Qual a representação que os professores têm a respeito da língua inglesa? Qual a representação dos professores a respeito de si mesmos como falantes da língua inglesa? Como essas representações dos professores de inglês influenciam na construção de suas identidades? As representações analisadas indicam como são construídas as identidades dos professores participantes enquanto falantes da língua inglesa

Made available in DSpace on 2016-07-27T14:19:53Z (GMT). No. of bitstreams: 1 DALVA DE JESUS CUTRIM MACHADO.pdf: 420046 bytes, checksum: 816e073b99d0b74ec40354c6ade154af (MD5) Previous issue date: 2012-01-10 / This study aimed to investigate the relationships between environmental variables and behavioral analysis that would promote relations analysis between speaker and listener. It also sought to investigate verbal and nonverbal context. From this perspective four experiments were performed: Experiment I had the following objectives: a) check if there was visual contact control variable (CV) and statement with the head (AC), which reinforce the verbal behavior of the speaker and listener in the total verbal episode b) consider whether different audiences control verbal differentiated responses between the behavior of the speaker and listener in a total verbal episode. Eight people aged 20 to 27 years of age took part of this experiment. The results showed that the control of the variable, visual contact had occured (VC) for most participants, while the variable (AC) there was no control for most participants. Experiment II aimed to systematically replicate the procedure adopted in the study of Simonassi, Tizo, Gomes and Alvarenga (2009) to see if: a) verbal contexts (instructions) and nonverbal (object) control on verbal responses in a total verbal episode b) whether the addition of a new object when presented in the presence of other existing objects controls verbal responses in a total verbal episode. This experiment was conducted with eleven participants aged 18 to 23 years. The results were similar to the experiment cited, and there was verbal control of responses when a new object was introduced: the answer had changed. Experiment III aimed to investigate: a) to test more systematically the possible influence on changing the verbal context (instructions) and verify if it would influence the control of verbal responses in a total verbal episode while maintaining the same nonverbal objects from Experiment II. Experiment III was conducted with eleven participants aged 19 to 25 years. The results showed that the change in the instructions had changed the context when compared to Experiment II. Experiment IV aimed to verify: a) what the written accounts of the participants were like with the same verbal context (instruction) from Experiment III and various contextual objects. This experiment was conducted with eleven participants aged 17 to 25 years. It was found that there was a significant reduction of responses when a different object was presented and there were changes in participants' verbal reports. It was concluded that both the verbal context (instruction) and the non-verbal context (objects) when they were changed, changed the response of the participants in the various experimental conditions. / O presente trabalho teve como objetivo investigar as relações entre as variáveis ambientais e comportamentais que favorecessem a análise das relações entre falante e ouvinte em um episódio verbal total. Também procurou investigar contexto verbal e não verbal. Foram realizados quatro Experimentos: o Experimento I teve como objetivo investigar: a) se houve ocorrência da variável contato visual (CV) e asserção com a cabeça (AC), que reforçam o comportamento verbal do falante no episódio verbal total; b) analisar se audiências diferenciadas controlam respostas verbais diferenciadas no comportamento do falante em um episódio verbal total. Participaram desse experimento, oito alunos com idades entre 20 a 27 anos de idade. Os resultados mostraram que houve controle da variável contato visual (CV) para a maioria dos participantes, enquanto a variável (AC) não houve controle da variável para a maioria dos participantes. Experimento II objetivou replicar de forma sistemática o procedimento adotado no estudo de Simonassi, Tizo, Gomes e Alvarenga (2010) com acréscimo de dois novos objetos, um lápis e uma bola, para verificar: 1) se contextos verbais (instruções) e não verbais (objetos) exercem controle sobre respostas verbais em um episódio verbal total; 2) se o acréscimo de novos objetos quando apresentados na presença dos outros objetos já existentes controlam respostas verbais em um episódio verbal total. Este experimento foi realizado com onze participantes com idades entre 18 a 23 anos. Os resultados foram semelhantes aos do Experimento citado, e houve controle das respostas verbais quando introduziu-se um novo objeto: o responder se modificou. Objetivo do Experimento III foi investigar sistematicamente a possível influencia na mudança dos comportamentos verbais que ocorreram no Experimento II, ou seja, se a modificação de estímulos verbais sob a forma de instrução em um contexto não verbal influenciaria o controle das respostas verbais em um episódio verbal total. Foi realizado com onze participantes com idade entre 19 a 25 anos. Os resultados mostraram que a mudança na instrução alterou o contexto quando comparado ao Experimento II. Experimento IV teve como finalidade investigar sistematicamente uma possível influencia na mudança dos relatos escritos dos participantes. Com o mesmo contexto verbal (instruções) semelhante ao do Experimento III. E a modificação de objetos não verbais. Este experimento foi realizado com onze participantes com idade entre 17 a 25 anos. Verificou-se que houve diminuição significativa das respostas quando o objeto diferente foi apresentado e houve alteração nos relatos verbais dos participantes. Concluiu-se que tanto o contexto verbal (instrução) quanto o contexto não verbal (objetos) quando foram alterados alteravam o responder dos participantes nas diversas condições experimentais.