Search results

International audience This article discusses the prosodic characteristics of French, in light of a metrical and perceptual analysis of various speaking styles. Our results lead us to propose the level of the prosodic word (pw) as the basic level of representation of French accentuation. This strong proposition, which has never been empirically attested to date, is based on two major results: listeners can perceive the final accent independently from intonation boundaries, at all levels of constituency; the processing of accentuation, particularly the initial accent, takes place at the lowest level of prosodic constituency (pw). The surface bipolar accentual demarcation of the lexical word invites us to reinterpret the status of accentuation in French and to question the notion of stress deafness in French. ; Nous proposons de discuter des caractéristiques prosodiques du français sur la base d'une analyse métrique et perceptive de corpus de parole variés. Nos résultats nous conduisent à proposer le niveau du mot prosodique (pw) comme niveau de représentation de l'accentuation du français. Cette proposition forte, jusque-là jamais attestée empiriquement, est fondée sur deux résultats majeurs : la capacité des sujets à percevoir une proéminence finale indépendamment de la frontière intonative, à tous les niveaux de constituance ; la perception et le traitement de l'accentuation, notamment initiale, au niveau le plus bas de la constituance prosodique (pw). La démarcation accentuelle bipolaire du mot lexical en surface nous invite à réinterpréter le statut de l'accentuation en français et à mettre en doute la notion de surdité accentuelle en français.

International audience ; This paper addresses the problem of the enrichment of transcriptions in the perspective of an automatic phonetization. Phonetization is the process of representing sounds with phonetic signs. There are two general ways to construct a phonetization process: rule based systems (with rules based on inference approaches or proposed by expert linguists) and dictionary based solutions which consist in storing a maximum of phonological knowledge in a lexicon. In both cases, phonetization is based on a manual transcription. Such a transcription is established on the basis of conventions that can differ depending on their working out context. This present study focuses on three different enrichments of such a transcription. Evaluations compare phonetizations obtained from automatic systems to a reference phonetized manually. The test corpus is made of three types of speech in French: conversational speech, read speech and political debate. A specific algorithm for the rule-based system is proposed to deal with enrichments. The final system obtained a phonetization of about 95.2% correct (from 3.7% to 5.6% error rates depending on the corpus).

International audience ; Notre objectif vise à estimer l'influence de différents niveaux d'enrichissement de la transcription sur l'étape de phonétisation de l'oral. Cette étude a été réalisée sur un corpus test de 7 minutes, réparties entre trois types de données différentes (parole conversationnelle spontanée, lecture et discours politique). Les résultats montrent que plus la transcription bénéficie d'enrichissements, meilleure est la phonétisation obtenue, quel que soit le type de corpus.

International audience ; This paper addresses the problem of the enrichment of transcriptions in the perspective of an automatic phonetization. Phonetization is the process of representing sounds with phonetic signs. There are two general ways to construct a phonetization process: rule based systems (with rules based on inference approaches or proposed by expert linguists) and dictionary based solutions which consist in storing a maximum of phonological knowledge in a lexicon. In both cases, phonetization is based on a manual transcription. Such a transcription is established on the basis of conventions that can differ depending on their working out context. This present study focuses on three different enrichments of such a transcription. Evaluations compare phonetizations obtained from automatic systems to a reference phonetized manually. The test corpus is made of three types of speech in French: conversational speech, read speech and political debate. A specific algorithm for the rule-based system is proposed to deal with enrichments. The final system obtained a phonetization of about 95.2% correct (from 3.7% to 5.6% error rates depending on the corpus).

International audience ; This paper addresses the problem of the enrichment of transcriptions in the perspective of an automatic phonetization. Phonetization is the process of representing sounds with phonetic signs. There are two general ways to construct a phonetization process: rule based systems (with rules based on inference approaches or proposed by expert linguists) and dictionary based solutions which consist in storing a maximum of phonological knowledge in a lexicon. In both cases, phonetization is based on a manual transcription. Such a transcription is established on the basis of conventions that can differ depending on their working out context. This present study focuses on three different enrichments of such a transcription. Evaluations compare phonetizations obtained from automatic systems to a reference phonetized manually. The test corpus is made of three types of speech in French: conversational speech, read speech and political debate. A specific algorithm for the rule-based system is proposed to deal with enrichments. The final system obtained a phonetization of about 95.2% correct (from 3.7% to 5.6% error rates depending on the corpus).

International audience ; Notre objectif vise à estimer l'influence de différents niveaux d'enrichissement de la transcription sur l'étape de phonétisation de l''oral. Cette étude a été réalisée sur un corpus test de 7 minutes, réparties entre trois types de données différentes (parole conversationnelle spontanée, lecture et discours politique). Les résultats montrent que plus la transcription bénéficie d'enrichissements, meilleure est la phonétisation obtenue, quel que soit le type de corpus.

International audience ; Notre objectif vise à estimer l'influence de différents niveaux d'enrichissement de la transcription sur l'étape de phonétisation de l'oral. Cette étude a été réalisée sur un corpus test de 7 minutes, réparties entre trois types de données différentes (parole conversationnelle spontanée, lecture et discours politique). Les résultats montrent que plus la transcription bénéficie d'enrichissements, meilleure est la phonétisation obtenue, quel que soit le type de corpus.

the analysis of language and speech is based on the study of a variety of areas ranging from phonetic to pragmatic, while taking into account the way in which they are expressed. For modern languages, while each area has a certain level of autonomy, it can only be explained in its interaction with others: linguistic information is the product of the convergence of multiple sources of information, which is also constrained by the environment in which the message was produced. Language analysis can therefore only be carried out taking into account the different ways in which information is expressed. However, to date, there is no integrated theory dealing with information from the different levels of the language chain, which is by its very nature a multi-modal sensory content. Although many annotation projects have developed in recent years, problems remain: annotated multimodal resources are hardly available for French; annotation standards do not fully meet needs and tools are not adapted. However, the development of this type of resources is useful in several respects, including the description of the information in each of the areas and their interactions. We present the Corpus of Interactional Data (CID), an audio and video corpus with 8 hours of dilogary in French, the annotations under consideration (phonetically, prosodic, morphosyntactical, mimo-gestual) and their state of implementation. ; L'analyse du langage et de la parole repose sur l'étude de domaines variés allant de la phonétique à la pragmatique, tout en prenant en compte les modalités de leur expression. Pour la linguistique moderne, si chaque domaine dispose d'un certain niveau d'autonomie, il ne peut être expliqué que dans son interaction avec les autres : l'information linguistique est le produit de la convergence de multiples sources d'information, information aussi contrainte par l'environnement où le message a été produit. L'analyse linguistique ne peut donc se faire qu'en tenant compte des différentes modalités d'expression de ...