Dominic Watt (University of York)
3 juin 2019, 14h au Laboratoire Parole et Langage
Sounds familiar! Identifying individual talkers is easier if they have the same local accent as the listener
Accent familiarity is known to affect the accuracy with which listeners identify individual talkers in speaker identification tasks. Speakers with familiar regional accents are identified more reliably by listeners than are speakers who use less familiar accents (the so-called ‘other-accent effect’), in the same way that listeners are poorer at identifying speakers of unfamiliar foreign languages than familiar ones (the ‘language familiarity effect’). In this talk, I describe an experiment designed to investigate whether the other-accent effect is observed when listeners are asked to identify speakers of localised subvarieties of a regional accent, that of the North-East of England (Braun et al. 2018). Listeners from the cities of Newcastle upon Tyne, Sunderland and Middlesbrough heard one of three target-present voice parades composed of samples of the speech of people from the same three locales. Target speakers from the same city as the listener were identified correctly significantly more often than were speakers of the other two accents, revealing that the ‘other-accent’ effect holds even across very small geographical distances (Newcastle and Sunderland are only 17km apart, for instance). I discuss the implications of this finding for the construction of voice parades in forensic cases where the witness has a different local accent from that of the suspect and foils whose speech is represented in the parade.
Braun, A., Llamas, C., Watt, D., French, P. & Robertson, D. (2018). Sub-regional ‘other-accent’ effects on lay listeners’ speaker identification abilities: A voice line-up study with speakers and listeners from the North East of England. International Journal of Speech, Language and the Law 25(2): 231-255.
Séminaire inter-équipe Interactions & REaDY
Vendredi 28 juin 2019
LPL – Salle de conférence B011
9h30 – A. Michelas & M. Champagne Lavau : Est-ce qu’on tient compte de notre interlocuteur dans nos choix linguistiques ? Le cas du focus contrastif en français
10h00 – N. Gala & C. David : De la complexité des textes à sa transposition didactique dans une classe de FLE multilingue : travaux exploratoires
10h45 – Pause-café
11h15 – K. Strijkers & N. Nguyen : Words in the speaker and listener’s mind: from individual to social level
[Réservé aux membres du LPL]
12h00 – Bilan annuel du LPL
12h30 – Déjeuner
Modélisation du dialogue: contrôle du dialogue et corpus multilingues
Pierre Lison (Norwegian Computing Center)
Mercredi 22 mai, 10h
Laboratoire Parole et Langage
Dans le cadre du PHC Aurora.
Contact pour le laboratoire : Laurent Prévot
A glance through Conversational agents
Mercredi 22 mai 11:30
Lina Maria Rojas Barahona, Orange Labs, Lannion
Conversational agents are once and again gaining great interest from academics and industrials. The availability of big data as well as the advances in processing units have made deep learning approaches feasible and promising, reviving the dream of creating artificial agents that can easily converse to people. Several solutions have been proposed since the first psychoanalyst chatbot (1966). From regular expressions and symbolic approaches (formal grammars and formal logics) to statistical approaches (probabilistic models and data-driven techniques, e.g. machine learning or deep learning). We have already obtained promising results. For instance, we know that machines can learn optimal strategies for simple tasks in small domains (task-oriented dialogue systems). Moreover, we are treating open domain dialogues by asking questions to online encyclopedias such as Wikipedia (conversational reading comprehension) and we are able to predict the best answer in chitchats (end to end neural approaches). In this talk I will review briefly the state of the art and highlight the open research problems on conversational agents.
Contact pour le laboratoire : Noël Nguyen