abstract: This study examined the development and acquisition of second language (L2) sounds by adult students enrolled in a communicative language program. The investigation explored the acquisition of L2 phones by analyzing the voice onset time (VOT) of word-initial voiceless stops in Spanish by native English speakers. A total of 40 subjects participated in the study and were divided into three groups; one group of students enrolled in a first semester course, another group of students enrolled in a third semester course, and the last group enrolled in a fifth semester course. The duration of VOT was compared between groups reading from a word list consisting of 60 words during the 13th to 15th weeks of the semester. Significant differences in VOT were found between the first and fifth semester groups, as well as the third and fifth semester groups suggesting that accurate acquisition of L2 phones and the formation of new phonetic categories are possible for late L2 learners in accordance with the Speech Learning Model. / Dissertation/Thesis / Masters Thesis Spanish 2018

Made available in DSpace on 2016-03-22T17:26:32Z (GMT). No. of bitstreams: 1 karolo.pdf: 888681 bytes, checksum: 0108eba901cb8567f6979752c9eb7dd3 (MD5) Previous issue date: 2011-03-21 / In this study, we investigated the phenomenon of aspiration of the voiceless plosives /p/, /t/ and /k/, in word-initial position, which constitutes one of the biggest difficulties of Brazilian students learning English. This analysis was made departing from two perspectives of data description: the binary one, which characterizes segments as aspirated or unaspirated , and the description in terms of the average VOT values of the segments produced. This distinction may lead to possible differences concerning the understanding of the developmental stages Brazilian learners of English go through, which may contribute to the understanding of L2 acquisition models and to the formalization in Phonological Theory. Teachers and students from an English course, as well as undergraduate students from Language and Literature major at Fundação Universidade Federal do Rio Grande (FURG), all of them from the Southern Brazilian city of Rio Grande, participated in this study. Participants were submitted to a placement test. The collection of data was made based on two instruments, containing words in Portuguese and English, having the segments /p/, /t/ and /k/ in initial position, with words that allow for the verification of quality effects of the following vowel. The conclusion is that, as well as in the study developed by Yavas & Wildermuth s (2006), the variables of place of articulation, height of the following vowel, number of syllables of the target-word and proficiency level of the learners have effects on the rates of aspiration/VOT produced by learners. This characterizes a very important discovery, mainly for those who are involved with the teaching of the L2 phonetic-phonological component. Furthermore, the description of VOT values proved able to highlight facts that were not clear under the binary approach, as it was able to reveal the gradience which characterizes the L2 acquisition process. In this sense, even though the target VOT values had not been completely reached, this latter approach was able to show that learners are already moving towards the target form, as they no longer produce the L1 pattern / Neste trabalho, foi investigado o fenômeno da aspiração das plosivas surdas /p/, /t/ e /k/, em posição inicial de palavras, o que constitui uma das dificuldades do aprendiz brasileiro ao aprender a língua inglesa. Tal análise foi feita a partir de duas perspectivas de descrição dos dados: a binária, com caracterização da presença/ausência de aspiração, e a descrição em termos de índices numéricos, referente às médias de VOT dos segmentos produzidos. A importância dessa distinção dá-se devido às possíveis diferenças no que concerne ao entendimento acerca dos estágios desenvolvimentais pelos quais passam os aprendizes, podendo contribuir com o entendimento dos modelos de aquisição de L2 e de formalização em Teoria Fonológica. Participaram deste estudo professores e estudantes de um curso de idiomas e acadêmicos do curso de Letras da Fundação Federal do Rio Grande (FURG), ambos da cidade do Rio Grande, os quais foram submetidos a um teste de nivelamento, a fim de homogeneizar a pesquisa. A coleta de dados foi feita com base em um instrumento de coleta de palavras em língua portuguesa, contendo os segmentos / p /, / t / e / k / em posição inicial; em um instrumento de coleta de palavras em língua inglesa, também contendo os segmentos / p /, / t / e / k / em posição inicial. Conclui-se que , bem como no estudo de Yavas & Wildermuth (2006), as variáveis referentes ao ponto de articulação, à altura da vogal seguinte, ao número de sílabas da palavra-alvo e ao nível de proficiência dos aprendizes exercem, sim, efeitos nos índices de aspiração/VOT produzidos pelos aprendizes, o que caracteriza uma descoberta bastante importante, sobretudo, para aqueles envolvidos com o ensino do componente fonético-fonológico da L2. Outrossim, a perspectiva de médias de VOT conseguiu evidenciar fatos que não se mostraram claros sob a perspectiva binária, pois elucidou a gradiência a qual caracteriza o processo de aquisição de L2, afinal, embora o alvo ainda não tenha sido plenamente atingido, desde o nível Básico de proficiência, os aprendizes pesquisados já se mostram em direção à forma-alvo, não mais produzindo os padrões da L1

Voice onset time (VOT) är en akustisk tidsparameter som återspeglar den talmotoriska samordningen. VOT betraktas även vara det mest tillförlitliga sättet att särskilja mellan tonande och tonlösa klusiler. VOT hos barn har tidigare studerats i logopediuppsatser vid Linköpings universitet (Lundeborg et al., 2012; Larsson & Wiman, 2011). I dessa arbeten har dock inga upprepade mätningar gjorts, varför det föreligger behov att fastställa hur konsistent barns VOT är. Denna studie syftar till att undersöka hur konsistent klusilproduktionen är hos svenska barn i femårsåldern, samt utreda förekomsten av eventuella könskillnader. I studien medverkade 31 barn, 13 flickor och 18 pojkar där medelåldern var 5:6 år. Barnen fick benämna bilder av minimala par med klusilord tre gånger. Samtliga testningar ljudinspelades. Inspelningarna analyserades därefter. Resultaten visade att barnen var förhållandevis konsekventa i VOT-värden vid de flesta klusilproduktionerna, variabilitet förekom mellan specifika klusilproduktionsomgångar. Bland de barn som hade förton vid produktionen av tonande klusiler noterades variabilitet i såväl förekomst som värde. Inga statistiskt signifikanta könsskillnader noterades.

AN ABSTRACT ON THE THESIS OFJose Benavides Pantoja, for the Master of Arts degree in Linguistics presented on April 30, 2020, at Southern Illinois University Carbondale. TITLE: PHONETIC VARIATION IN KAMËNTS̈Á: AN ANALYSIS OF STOPS AND AFFRICATESMAJOR PROFESSOR: Dr. Karen Baertsch This study contributes to the limited body of research concerned with documenting and studying the minority and indigenous languages of the world. More specifically, it focuses on Kamënts̈á, a language isolate spoken by less than 500 people in southern Colombia. According to UNESCO (2003), this language has been categorized as “definitely endangered” which makes its documentation even more critical. Although there have been initial investigations undertaken by Juajibioy, Howard, and McDowell, the scarce accessible research has relied on purely impressionistic reviews of the language, leading to unreliable conclusions and descriptions of the language. This thesis constitutes the first phonetic acoustic analysis of Kamënts̈á language in order to substantiate what has been described in the current literature. Additionally, the geographic location where Kamënts̈á emerged, together with Spanish, and Inga Kichwa has proven to be an intriguing sociohistorical and linguistic landscape for the study of these languages being an area worth of investigating. Given that there is no existing documentation of language variation in the language, the primary goal of this thesis is to analyze the production of plosive and affricate segments at a word-level to examine language variation at a phonetic level. Considering that Kamënts̈á has been poorly documented, it was not surprising for differences in the production to be found. One difference fell in the phonemic status of the segment between younger and older speakers of the language. The production of the segments in three linguistic environments at word-level attested for variation of some phonemes. For this study, data were collected from two experimental tasks. For the first task, eight Kamënts̈á-Spanish bilinguals (N=8) were recorded producing words in isolation in a word-elicitation exercise. The purpose of this task was to gather raw data on the production of the segments at word level. For the second task, the consultants were asked to take a sociolinguistic questionnaire. This instrument was designed to account for the linguistic ethnography and language ecology of the Kamënts̈á. The purpose of this task was also to tie sociolinguistic data from the speakers with any phonemic differences in the realization of the segments. Results supported the VOT framework proposed by Lisker and Abramson (1964) as an effective metric to measure the voicing status of plosives. For the analysis of affricates, the center of gravity and spectral peak location demonstrated accurate acoustic correlates to capture frication in the phonemes. Equally, amplitude provided valuable information for the analysis of prenasalization. What is more, consistent resemblances were found comparing Kamënts̈á with other studies regarding the acoustic cues, place and manner of articulation, and voicing status of the phonemes.

Made available in DSpace on 2016-03-22T17:26:23Z (GMT). No. of bitstreams: 1 Marta_Tessmann_Bandeira.pdf: 1541902 bytes, checksum: 7e4daab9f7d4916b9f64e370ac42d3fc (MD5) Previous issue date: 2010-03-23 / Arroio do Padre - RS was colonized by German and Pomeranians immigrants and is characterized as a geographic region relatively limited, which contributes to the inhabitants of the city to continue using the daily Pomeranian and / or German as their first language, a fact that gives rise to the bi and multilingualism. A multilingual person is defined as someone who can communicate in three or more languages with some degree of proficiency (Bialystok, 2001). The objectives of this study are: 1) examine the differences in the patterns of VOT in voiceless consonants of English produced by monolingual and multilingual children, attending the third grade in a municipal school in Arroio do Padre learning the English language in relation to the role of language Native (PB & Pomeranian, respectively), 2) compare the performance in executive functions (inhibitory control and attention) of mono- and multilingual children in accuracy and reaction time of a non-verbal task (Simon task) 3 ) to verify the performance differences in executive functions (inhibitory control and attention) between mono-and multilingual children in the accuracy and reaction time of a verbal task (Stroop test). To achieve the first objective, we used a tool for storytelling, in three languages (Portuguese, Pomerania and English), together with a game that would elicit the production of target words initiated by plosive voiceless. These words were recorded in studio and acoustically analyzed for the average VOT produced were compared. There was great difference between the averages for monolingual and multilingual production of plosives in English, indicating strong influence from the first language of the participants in the production of aspirated stops of English as L3 (in the case of bi / multilingual) and as L2 (in the case of monolingual). To test the executive functions, the same group of participants (20 monolingual and bilingual 20) underwent the Simon task and the Stroop test. We analyzed the results of reaction time and accuracy on each task, and found significant differences in all conditions of the two tests of the accuracy, and in most conditions, when measured reaction time. The results, which suggest that people develop the multi-processing related to executive functions faster and with higher levels of accuracy than monolingual, are discussed in light of the inhibitory control model (Green, 1998) and the theory of dynamical systems / O município de Arroio do Padre RS foi colonizado por imigrantes pomeranos e alemães e caracteriza-se como uma região geográfica de acesso relativamente limitado, fato que contribui para que os habitantes desse município continuem usando cotidianamente o pomerano e/ou o alemão como sua primeira língua, fato que enseja o bi e o multilinguismo. Define-se multilíngue como alguém capaz de se comunicar em três ou mais línguas com certo grau de proficiência (BIALYSTOK, 2001). Este trabalho investiga diferenças entre multi e monolíngues em tarefas envolvendo a produção de plosivas surdas de três línguas e funções executivas verbais e não verbais. Assim, os objetivos são: 1) analisar as diferenças relativas aos padrões de VOT em plosivas surdas do inglês produzidas por crianças monolíngues e multilíngues, cursando a terceira série numa escola municipal de Arroio do Padre aprendendo a língua inglesa, em relação ao papel da língua materna (PB e pomerano, respectivamente); 2) comparar o desempenho, em funções executivas (controle inibitório e atenção), de crianças mono e multilíngues, na acurácia e no tempo de reação de uma tarefa não verbal (tarefa de Simon); 3) verificar as diferenças de desempenho em funções executivas (controle inibitório e atenção) entre crianças mono e multilíngues na acurácia e no tempo de reação de uma tarefa verbal (Stroop test). Para a consecução do primeiro objetivo, foi utilizado um instrumento de contação de histórias, em três línguas (português, pomerano e inglês), acompanhado de um jogo que eliciava a produção de palavras-alvo iniciadas pelas consoantes plosivas surdas. Essas palavras foram gravadas em estúdio e analisadas acusticamente para que as médias de VOT produzidas fossem comparadas. Houve grande diferença entre as médias obtidas por multilíngues e monolíngues na produção de plosivas do inglês, apontando grande influência da primeira língua dos participantes na produção das oclusivas aspiradas do inglês como L3 (no caso dos bi/multilíngues) e como L2 (no caso dos monolíngues). Para testar as funções executivas, o mesmo grupo de participantes (20 monolíngues e 20 bilíngues) foi submetido à tarefa de Simon e ao teste de Stroop. Foram analisados os resultados do tempo de reação e acurácia em cada uma das tarefas, e encontraram-se diferenças significativas, em todas as condições dos dois testes relativas à acurácia, e na maioria das condições, quando se mediu tempo de reação. O resultados, que sugerem que pessoas multilíngues desenvolvem os processamentos ligados às funções executivas mais rapidamente e com níveis maiores de acurácia do que monolíngues, são discutidos à luz do modelo de Controle Inibitório (GREEN, 1998) e da Teoria dos Sistemas Dinâmicos

Madurese, a Western Malayo-Polynesian language spoken on the Indonesian island of Madura, exhibits a three-way laryngeal contrast distinguishing between voiced, voiceless unaspirated and voiceless aspirated stops and an unusual consonant-vowel (CV) co-occurrence restriction. The CV co-occurrence restriction is of phonological interest given the patterning of voiceless aspirated stops with voiced stops rather than with voiceless unaspirated stops, raising the question of what phonological feature they may share. Two features have been linked with the CV co-occurrence restriction: Advanced Tongue Root [ATR] and Lowered Larynx [LL]. However, as no evidence of voicing during closure for aspirated stops is observed and no other acoustic measures except voice onset time (VOT), fundamental frequency (F0), frequencies of the first (F1) and the second (F2) formants and closure duration relating to the proposed features have been conducted, it remains an open question which acoustic properties are shared by voiced and aspirated stops. Three main questions are addressed in the thesis. The first question is what acoustic properties voiced and voiceless aspirated stops share to the exclusion of voiceless unaspirated stops. The second question is whether [ATR] or [LL] accounts for the patterning together of voiceless aspirated stops with voiced stops. The third question is what the implications of the results are for a transparent phonetics-phonology mapping that expects phonological features to have phonetic correlates associated with them. In order to answer the questions, we looked into VOT, closure duration, F0, F1, F2 and a number of spectral measures, i.e. H1*-A1*, H1*-A2*, H1*-A3*, H1*-H2*, H2*-H4* and CPP. We recorded fifteen speakers of Madurese (8 females, 7 males) reading 188 disyllabic Madurese words embedded in a sentence frame. The results show that the three-way voicing categories in Madurese have different VOT values. The difference in VOT is robust between voiced stops on the one hand and voiceless unaspirated and voiceless aspirated stops on the other. Albeit statistically significant, the difference in VOT values between voiceless unaspirated and voiceless aspirated stops is relatively small. With regard to closure duration, we found that there is a difference between voiced stops on the one hand and voiceless unaspirated and aspirated stops on the other. We also found that female speakers distinguish F0 for the three categories while male speakers distinguish between F0 for voiced stops on the one hand and voiceless unaspirated and voiceless aspirated stops on the other. The results for spectral measures show that there are no significant differences in H1*-A1*, H1*-A3*, H1*-H2*, H2*-H4* and CPP between vowels adjacent to voiced and voiceless aspirated stops. In contrast, there are significant differences in these measures between vowels adjacent to voiced and voiceless unaspirated stops and between vowels adjacent to voiceless aspirated and voiceless unaspirated stops. Regarding the question whether voiced and voiceless aspirated stops share certain acoustic properties, our findings show that they do. The acoustic properties they share are H1*-A1* for both genders, H1*-H2* for females, H1*-A3* and H2*-H4* for males, and CPP for females at vowel onset and for males at vowel midpoint. However, they do not share such acoustic properties as VOT, closure duration and F0. Voiceless unaspirated and voiceless aspirated stops can be distinguished by VOT, F0 and spectral measures, i.e. H1*-A1*, H1*-A3*, H1*-H2*, H2*-H4* and CPP. However, these two voiceless stop categories have similar closure durations. As regards the question if [+ATR] or [+LL] might be responsible for the patterning together of voiceless aspirated stops with voiced stops, our findings suggest that either feature appears to be plausible. Acoustic evidence that lends support to the feature [+ATR] includes lower F1 and greater spectral tilt measures, i.e. H1*-A1*, H1*-A3*, H1*-H2* and H2*-H4*, and lower CPP values. Acoustic evidence that supports the feature [+LL] includes lower F1 and greater spectral tilt measures, i.e. H1*-A1*, H1*-A3*, H1*-H2* and H2*-H4*, and lower CPP values. However, the fact that voiceless aspirated stops are voiceless during closure raises a problem for the feature [+ATR] and the fact that F0 for voiceless aspirated stops is higher than for voiced stops also presents a problem for the feature [+LL]. The fact that not all acoustic measures fit in well with either feature is problematic to the idea that the relationship between phonetics and phonology is transparent in the sense that phonological features can be directly transformed into their phonetic correlates. Following the view that not all phonological features may not be expected to be phonetically grounded, for example, when they are related to historical sound change, we hold the idea of a phonetics-phonology mapping which allows for other non-phonetic factors to account for a phonological phenomenon. We also provide historical and loanword evidence which could support that voiceless aspirated stops in Madurese may have derived from earlier voiced stops, which probably retain their historical laryngeal contrast through phonologisation.

Submitted by Cristiane Chim (cristiane.chim@ucpel.edu.br) on 2017-12-19T10:56:48Z No. of bitstreams: 1 Tese Raquel Vaz Milheira.pdf: 1884697 bytes, checksum: 35ee30e38b6f26db68afe1d551d96ed3 (MD5) / Made available in DSpace on 2017-12-19T10:56:48Z (GMT). No. of bitstreams: 1 Tese Raquel Vaz Milheira.pdf: 1884697 bytes, checksum: 35ee30e38b6f26db68afe1d551d96ed3 (MD5) Previous issue date: 2017-05-22 / This study aims to investigate the role that VOT (Voice Onset Time) fulfills in process of acquisition of the voicing contrast between the stop segments that integrate the phonology of Brasilian Portuguese (PB). Children studied belong to a community in Rio Grande do Sul, Vista Alegre do Prata, which uses a variant of PB that still maintains contact with Polish and Italian, languages spoken by the older generations and that permeate the linguistic environment of family and regional contact. The research focuses on the VOT as a phonetic clue that can be used in the process of language acquisition, mainly for the establishment of the phonological value of the feature [voice], but which can also support the determination of the contrast between the place feature [ labial], [coronal] and [dorsal] in the class of stop consonants. In the discussion of the phonetic and phonological development process, through the analysis of VOT values in the class of stop consonants of PB the relationship between the linguistic forms of the children and those of their Caregivers is addressed. The corpus of this research consisted of linguistic data from four children aged 1: 4 to 2: 2, two boys and two girls; they are data of spontaneous speech that belong to the bank constituted by Souza (2015). From the corpus, all the target words that presented stop consonants were transcribed and submitted to acoustic analysis in the Praat program. The VOT values of each segment for each child were recorded, as well as the VOT values of the stop produced by their Caregivers. The analysis of the results showed that the children of Vista Alegre do Prata, in the process of acquiring Portuguese stops, suffer a preponderant influence of their Caregivers and that the VOT can be used as the primary clue for the acquisition of the phonological value of the feature [± voice] and can still be used as a secondary clue for the acquisition of place contrast in phonology of PB stop consonant class. Key words: Phonological acquisition; Portuguese stops; VOT; Acoustic analysis; [voicing] Feature; Place features. / Este estudo visa a investigar o papel que o VOT (Voice Onset Time) cumpre no processo de aquisição do contraste de vozeamento entre os segmentos plosivos que integram a fonologia do português brasileiro (PB). As crianças pesquisadas pertencem a uma comunidade do Rio Grande do Sul, Vista Alegre do Prata, que utiliza uma variante do PB que ainda mantém contato com o polonês e o italiano, línguas faladas pelas gerações de idade mais avançada e que permeiam o ambiente linguístico das crianças pelo contato familiar e regional. A investigação tem o foco no VOT como pista fonética que pode ser utilizada, no processo de aquisição da linguagem, principalmente para o estabelecimento do valor fonológico do traço [voz], mas que também pode subsidiar a determinação do contraste entre os traços de ponto [labial], [coronal] e [dorsal] na classe das consoantes plosivas. Na discussão do processo de desenvolvimento fonético e fonológico, por meio da análise dos valores de VOT na classe das plosivas do PB, é abordada a relação entre as formas linguísticas das crianças e as de seus Cuidadores. O corpus desta pesquisa foi constituído por dados linguísticos de quatro crianças acompanhadas desde 1:4 até 2:2 de idade, dois meninos e duas meninas; são dados de fala espontânea que pertencem ao banco constituído por Souza (2015). Do corpus, todas as palavras-alvo que apresentavam consoantes plosivas foram transcritas e submetidas à análise acústica no programa Praat. Foram registrados os valores de VOT de cada segmento para cada criança, assim como os valores de VOT das plosivas produzidas por seus Cuidadores. A análise dos resultados mostrou que as crianças de Vista Alegre do Prata, no processo de aquisição das plosivas do português, sofrem preponderante influência dos seus Cuidadores e que o VOT pode ser por elas utilizado como pista primária para a aquisição do valor fonológico do traço [±voz] e pode ainda ser utilizado como pista secundária para a aquisição do contraste de ponto na fonologia da classe das consoantes plosivas do PB.

abstract: ABSTRACT The absence of the consonant sound /p/ in Libyan Arabic leads Libyan speakers of English to pronounce /p/ as /b/. This study examines how Libyan Arabic speakers distinguish the English /p/ and /b/ in their production of L2 English. The study also examines the effect of the production contexts and the learning environment on two groups of Libyan Arabic speakers' attainment of the English /p/ in the USA and Libya. The study collected voice recordings of word-initial /p/ and /b/ in isolated-words, minimal pairs, and sentences in English from both Libyan Arabic speakers and American English speakers. The study also collected Libyan Arabic stop consonants /b/, /t/, /d/, /k/, and /g/ from the Libyan participants. The voice recording data were collected using the WhatsApp mobile application from all participants and the Libyan Arabic participants were also asked to fill an online survey. Using voice onset time (VOT) as a measurement tool, this study measured the English and Libyan Arabic data through Praat software. The findings show that most Libyan Arabic participants distinguish between /p/ and /b/, but they did not have as high VOT averages as the American participants' /p/. It also reveals that the production context, especially in minimal pairs and sentence contexts, has an effect on their participants' production. However, the learning environment does not have an effect on the Libyan participants' pronunciation of /p/ in this study. / Dissertation/Thesis / Masters Thesis English 2014

This study analyzes the production of Voice Onset Time (VOT) of natural Spanish-English bilinguals. VOT is a linguistic characteristic that measures the amount of aspiration occurring after the release of a stop consonant. In terms of VOT, English stop consonants differ substantially from their Spanish equivalents. This study analyzes whether or not natural bilinguals produce VOTs that approximate VOTs of monolingual speakers of each language. Participants completed two surveys to quantify their linguistic dominance in English and Spanish. They were then recorded performing similar speaking tasks in both languages. The conclusions show that natural bilinguals do not produce their English or Spanish VOTs within the monolingual norms defined in previous studies. If conclusions were to be drawn solely from this data, then the participants would theoretically have no monolingual-like language production of VOT. There is also no correlation between language dominance scores and production of VOT. These results support the conclusion that a natural bilingual is not the equivalent of two natural monolingual speakers. Significant correlations exist between VOT production and gender, age of learning English, and amount of time spent watching TV in each language. Another interesting conclusion is that many of the participants score more Spanish-dominant when a survey is given in Spanish and more English-dominant when the very same survey is given in English. This shows that even the language of a survey may skew responses slightly.

Cette thèse traite de l’acquisition des occlusives sourdes /p t k/ et sonores /b d g/ du français par 11 locutrices taïwanaises de niveau intermédiaire à avancé. La situation de Taïwan est qualifiée de diglossique, les locuteurs parlent généralement deux langues dont les principales sont le chinois mandarin et le taïwanais. Le chinois mandarin possède les occlusives /p t k ph th kh/ tandis que le taïwanais possède les occlusives /b g p t k ph th kh/. L’analyse acoustique des logatomes CVCVCVC où C=/b d g p t k/ et V=/a i u/ révèle une grande hétérogénéité entre les locutrices : les indices des natifs du français pour opposer entre sourdes et sonores sont parfois utilisés par les non-natifs, parfois non. On note l’influence du chinois mandarin. Un test de perception révèle une moins bonne discrimination des paires de consonnes /b p/, /d t/ et /g k/ en syllabe CV si V=/a/, comparé à /i u/. Ces résultats suggèrent une tendance générale chez ces auditrices à mieux discriminer les occlusives du français lorsque le VOT des sourdes est plus long et à ne pas tenir compte du VOT négatif des voisées. En perception, les indices pour discriminer les occlusives aspirées et non-aspirées du chinois mandarin semblent ainsi également être utilisés en français. Nous n’avons pas relevé de signe d’une influence du taïwanais, où l’opposition de voisement existe cependant. / This dissertation deals with the acquisition of French voiceless stops /p t k/ and voiced stops /b d g/ by 11 Taiwanese intermediate or advanced learners of L2 French. The linguistic situation in Taiwan is described as diglossia. Most speakers speak two languages, mainly Mandarin Chinese and Taiwanese. Mandarin Chinese has plosives /p t k ph th kh/ while Taiwanese has /b g p t k ph th kh/. An acoustic analysis of CVCVCVC logatoms where C = /b d g p t k/ and V = /a i u/ shows important heterogeneity among speakers. The cues used by French native speakers to oppose voiceless and voiced stops are irregularly used by non-native speakers. The influence of Mandarin Chinese is noted. A perception test shows poorer discrimination among pairs of consonants (/b p/, /d t/ and /g k/) in CV syllable when V = /a/, as compared to /i u/. The results show that non-native listeners tend to, firstly, better discriminate the voiceless plosives of French when the VOT is longer and secondly, ignore the negative VOT of voiced stops. As regards perception, the cues used in Mandarin Chinese to discriminate between aspirated and non-aspirated stops consonants seem to be used in French too. No clue to the influence of Taiwanese has been found, although the opposition of voicing exists.