Le but de cette thèse est de mieux comprendre le phénomène de renforcement initial pour les voyelles du français : ses mécanismes de réalisations et sa fonction linguistique. Pour cela, cette thèse compare la réalisation acoustique et articulatoire (lèvres et menton) des voyelles orales du français /i, e, ɛ, a, y, ø, œ, u, o, ɔ/ en position strictement initiale de Groupe Intonatif et en position médiane de Groupe Intonatif. Le corpus a été enregistré par quatre locutrices avec deux appareils en simultané : une caméra pour l’analyse de l’aperture et l’étirement des lèvres et un appareil de capture mouvements (Qualisys) pour l’analyse articulatoire de la protrusion des lèvres et de l’abaissement du menton. Une étude acoustique a également été menée (formants, intensité, durée). Cette thèse montre que, d’une part, le renforcement initial modifie des caractéristiques phonétiques des voyelles qui peuvent contribuer, à renforcer les contrastes syntagmatiques en augmentant la sonorité de la voyelle: augmentation de l’aperture et l’étirement aux lèvres pour l’ensemble des voyelles et augmentation d’intensité acoustique pour la plupart. D’autre part, le renforcement initial renforce des caractéristiques acoustiques et articulatoires propres à chaque voyelle. Ces variations contribuent à maximiser les contrastes paradigmatiques entre voyelles : soit par une maximisation de chacune des valeurs du contraste ([+F] et [-F]), soit par une maximisation d’une des deux valeurs du contraste ([+F] ou [-F]). / The goal of this thesis is to better understand the phenomenon of domain-initial strengthening on French vowels: how it is realized and its linguistic function. We compare the acoustic and articulatory variations (on the lips and chin) of the French oral vowels /i, e, ɛ, a, y, ø, œ, u, o, ɔ/ in initial position of Intonational Phrase and in medial position of Intonational Phrase. The corpus was recorded by four female speakers with two different systems simultaneously. Lip aperture and lip width were analyzed from the video data and lip protrusion and lowering of the chin were captured by an optical motion capture system (Qualisys). An acoustic study was also conducted (formants, intensity, duration). Results show that initial strengthening affects phonetic properties of vowels which may contribute to enhance syntagmatic contrasts by increasing their sonority: increased lip aperture and width for all vowels, and increased acoustic intensity for most. Moreover, initial strengthening enhances vowel-specific acoustic and articulatory properties. Consequently, initial strengthening contributes to maximize paradigmatic contrasts between vowels, either by maximizing contrast of each of the values ([+F] and [-F]), either, or by maximizing the value of a single contrast ([+F] or [-F]).

TITLE: Teaching of French pronunciation aimed at nasal vowels SUMMARY: The thesis deals with the issues of French pronunciation teaching at Czech schools. The theoretical part is aimed at teaching of pronunciation, the key concepts of phonetics and its importance in learning of foreign language. Then the thesis deals with the base of French phonetic system, especially the nasal vowels, the comparison of Czech and French phonetics and the main methods of teaching phonetics. The nasal vowels and their pronunciation are the starting point for the practical part - research in the field of the ability of Czech students to recognize and pronounce the nasal vowels in a good way. The research is based on 95 recordings of Czech students from different classes (from high school to bachelor's programme). KEYWORDS: French language, phonetics, phonology, teaching of phonetics, nasal vowels

Phonetic mergers in American English have been studied throughout the last half century. Previous research has contributed social and phonetic explanations to the understanding of front and back vowel mergers before /l/, front vowel mergers before nasals and phonetically unconditioned back vowel mergers. Using data from the Linguistic Atlas of the Gulf States (LAGS) and the Linguistic Atlas of the Middle and South Atlantic States (LAMSAS), this thesis examines the spread of the front vowel mergers in Mary and merry and the back vowel mergers in horse and hoarse. The two complementary sources of data allow for a social and phonetic approach to the examination of the merger.

The aim of this thesis is creation the appropriate multimedia support for signals and system with continuous time. The understanding of this issue is very important, because the obligatory subject Signals and systems, exactly BSIS, is taught at the EST bachelor degree. The understanding is also necessary prerequisite to successful understanding next topics in other related subjects. The next part of this thesis is focused on one dimension discrete signals. Concretely, the aim of this part is a realization of software system. Designed system has some basic operations (the signal energy, the number of signal zero crossing etc.) with sound files and also some advance functions e.g. vowel seeking and separating in fluent speech. The system is divided into two main parts. The first one analyzes sound files, creates the new sound file with wanted vowel and matrices with important parameters for other processing. The second program computes with given data, which statistically evaluates in other steps. The final system can be useful for speaker recognition, his emotional status etc.

This diploma thesis deals with the speech signal processing and vowel analysis mainly to uncover differences in speech features depending on the emotional state of speaker. Created application ARePa for speech signal processing was developer in Matlab environment and contains Graphical User Interface (GUI) for better manipulation with ARePa and analysed records. The application includes a complete analysis of the speech signal and further comparison of current feature with feature values from database using histograms. Of course, the developer application allows the archivation of currently analysed records into database.

The purpose of this study was to investigate the effects of speaker-sex-difference on listeners' perception of vocal roughness in the vowel [ӕ] produced by normal male and female speakers. In a previous investigation by Wendahl (1963) it was found that when listening to two synthesized vowels, of equal aperiodicity, judges tended to rate the lower pitched vowel as being more vocally rough. If this is true for listeners’ perception of human vowel productions as well then it might be advantageous for voice clinicians, when making vocal roughness assessments, to regard male and female speakers as two separate populations in view of the inherent pitch differences between the sexes. In this current investigation, pairs of vowels produced by normal adult male and female speakers were presented to 10 speech pathologists (5 males and 5 females). Each vowel pair contained one male and one female production of the vowel [ӕ] which had been assigned equal roughness ratings in a previous judging task. The 50 vowel pairs contained 10 pairs of vowels at each of five roughness rating levels. The 10 judges were required to listen to each of the 50 pairs and to make a forced choice selection of the most vocally rough production within each pair. The findings in this study revealed that for the 50 vowel pairs the judges selected the vowels produced by males as being more vocally rough a significantly greater proportion of the time. With respect to the five roughness rating levels, judges chose the male produced vowels as being rougher a significantly greater proportion of the time at rating levels one, three and five but illustrated no significant preference between the sexes at rating levels two and four. Further analysis revealed that the five male judges selected the vowels produced by males as being the rougher a significantly greater proportion of' the time for all 50 pairs at each of the five roughness rating levels. The five female judges, on the other hand, illustrated no significant preferences between the sexes for the 50 vowel pairs. They did show a significant preference for the males at rating level one, a significant preference for the females at rating level two but no significant at rating levels three, four and five. In addition, male judges illustrated substantially greater inter-judge agreement and intra-judge reliability for this judging task than did the female judges.

This study examines the effects of experience in German on the categorical perception of German rounded vowels, namely /uː/, /ʊ/, /oː/, /ɔ/, /yː/, /ʏ/, /øː/, and /œ/, by native speakers of American English (AE). Of special interest is whether more experience in German leads to more accurate perception of German front rounded vowels, namely /yː/, /ʏ/, /øː/, and /œ/, which do not have correlates in American English and are well known to cause perceptual problems for native AE speakers (Strange, Bohn, Trent, & Nishi, 2004). Subjects in this study were students at Brigham Young University that were divided into 4 experimental groups: students at the end of first-semester German with no residency in a German-speaking country (101 group); students at the end of third-semester German with no residency (201 group); students in third-year or higher German courses with less than 4 months of residency (300+ group); students in third-year or higher courses with 16 or more months of residency (300+Resi group). A control group of native German speakers also participated. Subjects completed a forced-choice identification task in which they selected the German word they thought they heard. The results of the task indicate that experience in German did affect native AE-speaking subjects' overall identification accuracy of German rounded vowels. In particular, a statistically significant difference was found between the 101 and 300+Resi groups for all German rounded vowels except /uː/ and /ʊ//, suggesting that experience significantly affected AE subjects' perception of all of these vowels except /uː/ and /ʊ/.

The current study examined the lingual kinematic and acoustic effects of bite blocks on vowels in a sentence context. Twenty adult native English speakers (10 male, 10 female) with no speech, language, or hearing deficits participated in the study. The corner vowels found in the sentence, The blue spot is on the black key again (i.e., /u/, /ɑ/, /æ/, /i/), were measured kinematically and acoustically immediately before and after bite block insertion. The participants' speech was audio-recorded and their lingual articulatory movements were measured with a Northern Digital Instruments Wave electromagnetic articulograph. The sensor coils were attached to three different parts on the tongue (back, middle, and front). Acoustic analysis of the vowel formants revealed that the vowel articulation index and vowel space area decreased significantly following bite block insertion. Kinematic analysis of the sensors on the tongue revealed that the kinematic vowel articulation index decreased significantly for the back and middle of the tongue but not for the front. Thus, adjustments to the position of the front of the tongue were sufficient to compensate for the bite block perturbation, while the same measures for the back and middle of the tongue were significantly affected. This was likely due to the relative independence in the movement of the front of the tongue, given its distance from the posterior point of attachment between the tongue and mandible. These findings suggest that the effects of articulatory perturbation can be more fully understood when kinematic and acoustic measures are considered together.

No description available.

No description available.