En traitement du signal audio, les plosives sont des sons de parole très importants au regard de l’intelligibilité et de la qualité. Les plosives sont cependant difficiles à modéliser à l’aide des techniques usuelles (prédiction linéaire et codage par transformée), à cause de leur dynamique propre importante et à cause de leur nature non prédictible. Cette étude présente un exemple de système complet capable de détecter, segmenter, et altérer les plosives dans un flux de parole. Ce système est utilisé afin de vérifier la validité de l’hypothèse suivante : La phase d’éclatement (de burst) des plosives peut être mise à zéro, de façon perceptuellement équivalente. L’impact sur la qualité subjective de cette transformation est évalué sur une banque de phrases enregistrées. Les résultats de cette altération hautement destructive des signaux tendent à montrer que l’impact perceptuel est mineur. Les implications de ces résultats pour le codage de la parole sont abordées.

abstract: This study analyzes the Spanish of native speakers from Northern Mexico in order to ascertain the presence of the voiced labiodental fricative segment [v] when the sound is orthographically represented with the letter `v'. The study examines some of the internal and external factors that predict the labiodental fricative pronunciation of the letter `v'. This study is based on the theoretical framework of phonology as described by Piñeros (2009) and Hualde (2005). The study examined all instances in the data when a sound is written with the letter `v' to investigate if the sound is pronounced as the faithful voiced plosive bilabial allophone [b] of the phoneme /b/, the spirantized allophone [β], or the voiced labiodental fricative allophone [v]. Four speakers, a male and a female with an incomplete secondary education, and a male and a female with a graduate level education participated in the study. All participants were interviewed for one hour, read a word list, and read a paragraph provided by the researcher. The researcher coded the data using the phonetic analysis software Praat and all data were statistically analyzed using the multivariate software analysis program Goldvarb X in order to investigate the presence of the voiced labiodental fricative allophone [v] and predict what internal and external factors most influence its production. From this study it is obvious that the most influential factor favoring the realization of the labiodental fricative allophone [v] is orthography. When the phonetic segment was represented with the grapheme , the phonetic realization was more likely to be the labiodental fricative [v]. The level of education of the speaker and the formality of the stylistic setting were also determined to be influential factors. Speakers with a higher degree of education and stylistic settings with a higher degree of formality favored the realization of the labiodental fricative [v]. With regards to the internal factors, rather than external factors, a preceding phonological segment of a vowel or fricative dental [s] also favored the realization of the labiodental fricative [v]. / Dissertation/Thesis / M.A. Spanish 2012

This thesis presents a study concerning stop consonant to vowel transitions which are modeled making use of acoustic tube model. Characteristics of the stop consonant to vowel transitions are tried to be obtained first. Therefore several transitions including fricative to vowel transitions are examined based on spectral and time related properties. In addition to these studies, x-ray snapshots, lip videos and also experiments including subjects are used to intensify the characterization, from the production and the perception side of views. As results of these studies the plosive to vowel transitions are observed to be uttered by exponential vocal tract movements and the perception mechanism is observed to be highly related with exponential spectral changes. A model, based on the acoustic tube model, is tried to be established using the knowledge and the experience gained during characterization therefore proposed model involves the vocal tract parameters observed in characterization part. Finally, plosive to vowel transitions including three types of plosives (alveolar, labial and velar) are synthesized by the proposed model. The formants of the synthesized sounds are compared with the formants of the natural sounds. Also the intelligibility tests of these sounds are done. Performance evaluation tests show the proposed model&rsquo / s performance to be satisfactory.

Made available in DSpace on 2016-04-28T18:23:33Z (GMT). No. of bitstreams: 1 Maria Teresa Rosangela Lofredo Bonatto.pdf: 2333969 bytes, checksum: 5ca789ba06460e596a8e60420ff59a97 (MD5) Previous issue date: 2007-09-13 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / Plosive sounds are produced early in childhood, yet it is not uncommon cases in speech therapy clinics of children who have difficulty pronouncing them, particularly as regards the distinction between voiced and voiceless plosives. Existing literature in Phonetics states that the production of plosives involves the constriction of the vocal folds which can in turn increase the degree of supra-laryngeal obstruction in par with that of sublaryngeal obstruction and, as a result, make voicing difficult. In order to produce voiced obstruent sounds similarly to adults, the child has to learn to synchronize glottal and oral gestures. Procedures for the acoustic investigation of such plosives abound in the literature, among which we can mention Voice Onset Time (VOT), the total duration of the plosive segment and the duration of the vowel sound which either precedes or follows it. The aim of the present study is to characterize, by means of acoustic phonetics investigation, the production of plosive consonants in Brazilian Portuguese (BP), in the speech of three and twelve year-old children. In order to undertake such investigation, we used a corpus comprised of six plosives of BP, extracted from the production of two-syllable words inserted in carrier phrase. The children were selected after detailed analysis and the acoustic sign inspection was carried out in the sound wave shape using the broadband spectrogram. Extractions were made of VOT as well as of each of the vowel and consonant segments from the /a/ sound of diga , besides the key-words, of the word baixinho and the full sentence, starting from the first vowel (ex: D/i/g/a/p/a/p/a/baixinho ). The corpus composed of 3-year-old children was used in the perceptive study with the participation of 120 judges. The results showed that the judges identified the plosives /b/, /d/ e /g/ as voiced when accompanied by at least 53% of sound spectra. As regards /p/, /t/ e /k/, plosives, the appearance of breathy vowel caused these consonants to identified as voiced, while aspiration had no effect whatsoever in the perception of judges in the case of voiceless plosives. As to production results, variability in the mean scores and high standard deviation for voiceless, tonic and post-tonic plosives were observed by and large in all the age brackets. Overall, VOT proved to be a satisfactory parameter for voicing differentiation in infant talk, observed in the speech of children over 3 years of age. However, VOT measures were not consistent enough to differentiate plosives, bilabials and alveolar sounds. This study adopted theoretical notions as proposed by Source - Filter Theory (Fant, 1960), Articulatory Phonology (FAR - Browman and Goldstein, 1986; 1990. 1992) and Articulatory Acoustic (FAAR - Albano, 2001), Direct Realism Theory (Fowler, 1986), and the theory of the phonological system acquisition termed by Albano (1990) as O Toque de Ouvido . Based on this theoretical background, we can affirm that voicing contrast was present in the speech of all the children. We can also affirm that in the production of 3-year-olds there have been greater latency between the coordination of glottal and articulatory gestures, which was observed by the great availability in the productions in the measures of VOT. Velar sounds showed less variability. It was inferred , based on the analysis of spectrographic characteristics along with theoretical foundations of gestural phonology (FAR and FAAR), that small children have difficulty interrupting a gesture and starting another one, probably due to difficulties arisen from overlapping gestures , which are better articulated the older one gets / Os sons obstruintes plosivos são produzidos logo cedo pelas crianças, mas apesar disso é comum encontrarmos na clínica fonoaudiológica aquelas que apresentam dificuldades para pronunciá-los, principalmente em relação a estabelecer a distinção entre os vozeados e os não-vozeados. Para a produção dos sons vozeados a literatura fonética explica que o grau de constrição das pregas vocais pode aumentar o nível de obstrução supra-laríngea igualando-o ao da sub-larínge e como conseqüência dificultar o vozeamento. Para produzir os sons obstruintes vozeados semelhantes aos do adulto, a criança tem que aprender a sincronizar o gesto glotal e oral. Para a investigação acústica da produção dessas plosivas a literatura aponta várias medidas, dentre elas o Voice Onset Time (VOT), a duração total do segmento correspondente à plosiva e a duração da vogal que a precede ou a sucede. O objetivo do presente estudo é caracterizar, por meio de uma investigação fonéticoacústica, a produção das consoantes plosivas no Português Brasileiro (PB), na fala de crianças de 3 a 12 anos. Para realizar tal investigação, utilizamos um corpus contendo as seis plosivas do PB, a partir da produção de palavras dissílabas, inseridas em frases-veículo. As crianças foram selecionadas após análise criteriosa e a inspeção do sinal acústico foi realizada na forma de onda e no espectrograma de banda larga. Foram realizadas a extração do VOT e de cada um dos segmentos vocálicos e consonantais a partir do /a/ de diga , além das palavras-chave, da palavra baixinho e da frase completa, começando com a primeira vogal (Ex: D/i/g/a/p/a/p/a/baixinho ). O corpus das crianças de 3 anos foi utilizado para a realização do estudo perceptivo com a participação 120 juizes. Quanto aos resultados obtidos, os juízes identificaram as plosivas /b/, /d/ e /g/ como vozeadas quando apresentavam, pelo menos, 53% de barra de sonoridade. No caso das plosivas /p/, /t/ e /k/, a presença da breathy vowel favoreceu a identificação dessas consoantes como vozeadas e a presença da aspiração não interferiu no julgamento dos juizes para as plosivas não-vozeadas. Quanto aos resultados da produção, pudemos observar para as plosivas não-vozeadas, tônicas e pós-tônicas, variabilidade nos valores das médias e desvios-padrão elevados para todas as faixas etárias. De forma geral, verificamos que o VOT foi um parâmetro satisfatório para a diferenciação do vozeamento na fala infantil, e esta foi encontrada a partir dos 3 anos. O VOT não foi específico para diferenciar entre o ponto articulatório das plosivas bilabiais e alveolares. Os pressupostos teóricos que adotamos, como a Teoria da Fonte e do Filtro (Fant, 1960), a Fonologia Articulatória (FAR - Browman e Goldstein, 1986; 1990; 1992) e a Acústica Articulatória (FAAR - Albano, 2001), a Teoria do Realismo Direto (Fowler, 1986) e a proposta de aquisição do sistema fonológico que Albano (1990) denominou O Toque de Ouvido , nos permitem afirmar que o contraste de vozeamento esteve presente na produção de fala de todas as crianças. Podemos afirmar, também, que principalmente na produção das crianças de 3 anos, ocorreu maior latência entre a coordenação dos gestos glotal e articulatório, que foi observada pela grande variabilidade em suas produções em termos de VOT, sendo que as velares revelaram menor variabilidade. Inferimos com base na análise das características espectrográficas e com apoio nos fundamentos teóricos da fonologia gestual (FAR e FAAR) que as crianças pequenas têm dificuldades em interromper um gesto e iniciar outro, provavelmente em função de dificuldades em relação à sobreposição de gestos e com o aumento da idade essa sobreposição passa a ser melhor realizada

Neuromuscular explosive strength (defined as rate of force development; RFD) is considered important during explosive functional human movements; however this association has been poorly documented. It is also unclear how different variants of strength training may influence RFD and its neuromuscular determinants. Furthermore, RFD has typically been measured in isometric situations, but how it is influenced by the types of contraction (isometric, concentric, eccentric) is unknown. This thesis compared neuromuscular function in explosive power athletes (athletes) and untrained controls, and assessed the relationship between RFD in isometric squats with sprint and jump performance. The athletes achieved a greater RFD normalised to maximum strength (+74%) during the initial phase of explosive contractions, due to greater agonist activation (+71%) in this time. Furthermore, there were strong correlations (r2 = 0.39) between normalised RFD in the initial phase of explosive squats and sprint performance, and between later phase absolute explosive force and jump height (r2 = 0.37), confirming an association between explosive athletic performance and RFD. This thesis also assessed the differential effects of short-term (4 weeks) training for maximum vs. explosive strength, and whilst the former increased maximum strength (+20%) it had no effect on RFD. In contrast explosive strength training improved explosive force production over short (first 50 ms; +70%) and long (>50 ms; +15%) time periods, due to improved agonist activation (+65%) and maximum strength (+11%), respectively. Explosive strength training therefore appears to have greater functional benefits than maximum strength training. Finally, the influence of contraction type on RFD was assessed, and the results provided unique evidence that explosive concentric contractions are 60% more effective at utilising the available force capacity of the muscle, that was explained by superior agonist activation. This work provides a comprehensive analysis of the association between athletic performance and RFD, the differential effects of maximum vs. explosive strength training, and the influence of contraction type on the capacity for RFD.

The aim of this diploma thesis is to closely investigate and describe selected phonetic properties of Czech oral plosives [p t ť k b d ď g] in normal speech. The re- search focuses mostly on temporal characteristics in various contexts; moreover, we deal with the possible manners of plosive articulation in Czech, including non-canonical realizations. Another important part of the present study is a comparison with earlier studies. In the domain of temporal characteristics we examine the influence of various contexts on phone duration. The contexts include: phonetic context, speaker gender, the position of the plosive in the stress unit or articulation rate. We also examine a possible connection between the duration of the plosive and its realization. In the part that deals with the manner of articulation we focus on the individual phonetic properties, for which we investigate especially the conditions and degree of their stability. Furthermore we describe the specific plosive articulations in certain pho- netic contexts and some alternative realizations of certain plosives. The speech material used is mostly spontaneous. All the speakers are non- professionals from various TV broadcasts. Keywords: plosives, duration, phone segmentation, phonetic properties, pho- netic context, gender, articulation...