Search

Your search keyword '"PPC"' showing total 673 results
Start Over Author "PPC"
673 results on '"PPC"'

107. Test and teach Number Ninety-Two: Part 1

Author

Kwan-Hon Chan, M.F. Cheung, Ppc Ip, and Wai-Kuen Ng

Subjects
Text mining, Computer science, business.industry, Artificial intelligence, business, computer.software_genre, computer, Natural language processing, Pathology and Forensic Medicine, Test (assessment)
Published
1999
Full Text
View/download PDF

110. IL-4 and IL-13 inhibit IL-1β and TNF-α induced kinin B1 and B2 receptors through a STAT6-dependent mechanism.

Author

Souza, PPC, Brechter, AB, Reis, RI, Costa, CAS, Lundberg, P, and Lerner, UH

Subjects
INTERLEUKINS, TUMOR necrosis factors, STAT proteins, BONE resorption, GENE expression, KININS, LABORATORY mice
Abstract

Background and Purpose Bone resorption induced by interleukin-1β ( IL-1β) and tumour necrosis factor ( TNF-α) is synergistically potentiated by kinins, partially due to enhanced kinin receptor expression. Inflammation-induced bone resorption can be impaired by IL-4 and IL-13. The aim was to investigate if expression of B1 and B2 kinin receptors can be affected by IL-4 and IL-13. Experimental Approach We examined effects in a human osteoblastic cell line ( MG-63), primary human gingival fibroblasts and mouse bones by IL-4 and IL-13 on mRNA and protein expression of the B1 and B2 kinin receptors. We also examined the role of STAT6 by RNA interference and using Stat6-/- mice. Key Results IL-4 and IL-13 decreased the mRNA expression of B1 and B2 kinin receptors induced by either IL-1β or TNF-α in MG-63 cells, intact mouse calvarial bones or primary human gingival fibroblasts. The burst of intracellular calcium induced by either bradykinin ( B2 agonist) or des- Arg10-Lys-bradykinin ( B1 agonist) in gingival fibroblasts pretreated with IL-1β was impaired by IL-4. Similarly, the increased binding of B1 and B2 ligands induced by IL-1β was decreased by IL-4. In calvarial bones from Stat6-deficient mice, and in fibroblasts in which STAT6 was knocked down by siRNA, the effect of IL-4 was decreased. Conclusions and Implications These data show, for the first time, that IL-4 and IL-13 decrease kinin receptors in a STAT6-dependent mechanism, which can be one important mechanism by which these cytokines exert their anti-inflammatory effects and impair bone resorption. [ABSTRACT FROM AUTHOR]

Published
2013
Full Text
View/download PDF

114. Co-ingestion of a protein hydrolysate and amino acid mixture with carbohydrate improves plasma glucose disposal in patients with type 2 diabetes.

Author

Manders RJF, Wagenmakers AJM, Koopman R, Zorenc AHG, Menheere PPC, Schaper NC, Saris WHM, and van Loon LJC

Abstract

BACKGROUND: Although insulin secretion after carbohydrate ingestion is severely impaired in patients with type 2 diabetes, amino acid and protein co-ingestion can substantially increase plasma insulin responses. OBJECTIVE: We investigated insulin responses and the subsequent plasma glucose disposal rates after the ingestion of carbohydrate alone (CHO) or with a protein hydrolysate and amino acid mixture (CHO+PRO) in patients with a long-term diagnosis of type 2 diabetes. DESIGN: Ten type 2 diabetic patients [mean (+/-SEM) age: 62 +/- 2 y; body mass index (kg/m(2)): 27 +/- 1] and 9 healthy control subjects (age: 58 +/- 1 y; body mass index: 27 +/- 1) participated in 2 trials in which the plasma insulin response was measured after the ingestion of 0.7 g carbohydrate x kg(-1) x h(-1) with or without 0.35 g x kg(-1) x h(-1) of a mixture that contained a protein hydrolysate, leucine, and phenylalanine. Continuous infusions with [6,6-(2)H(2)]glucose were then given to investigate plasma glucose disposal. RESULTS: Plasma insulin responses were higher by 299 +/- 64% and 132 +/- 63% in the CHO+PRO trial than in the CHO trial in the diabetic patients and the matched control subjects, respectively (P < 0.001). The subsequent plasma glucose responses were reduced by 28 +/- 6% and 33 +/- 3% in the CHO+PRO trial than in the CHO trial in the diabetic patients and the matched control subjects, respectively (P < 0.001). The reduced plasma glucose response in the diabetic patients was attributed to a 13 +/- 3% increase in glucose disposal (P < 0.01). CONCLUSIONS: The combined ingestion of carbohydrate with a protein hydrolysate and amino acid mixture significantly increases de novo insulin production in patients with a long-term diagnosis of type 2 diabetes. The increased insulin response stimulates plasma glucose disposal and reduces postprandial glucose concentrations. Copyright © 2005 American Society for Clinical Nutrition [ABSTRACT FROM AUTHOR]

Published
2005

116. Gamma-linolenic acid supplementation for prophylaxis of atopic dermatitis -- a randomized controlled trial in infants at high familial risk.

Author

van Gool CJA, Thijs C, Henquet CJM, van Houwelingen AC, Dagnelie PC, Schrander J, Menheere PPC, and van den Brandt PA

Abstract

BACKGROUND: Studies suggest that low concentrations of n-6 long-chain polyenes in early life are correlated to atopic disease in later life. OBJECTIVE: The purpose of the study was to investigate the possible preventive effect of gamma-linolenic acid (GLA) supplementation on the development of atopic dermatitis in infants at risk. DESIGN: In a double-blind, randomized, placebo-controlled trial, formula-fed infants (n = 118) with a maternal history of atopic disease received borage oil supplement (containing 100 mg GLA) or sunflower oil supplement as a placebo daily for the first 6 mo of life. Main outcome measures were the incidence of atopic dermatitis in the first year of life (by UK Working Party criteria), the severity of atopic dermatitis (SCORing Atopic Dermatitis; SCORAD), and the total serum immunoglobulin E (IgE) concentration at the age of 1 y. RESULTS: The intention-to-treat analysis showed a favorable trend for severity of atopic dermatitis associated with GLA supplementation ( x+/- SD SCORAD: 6.32 +/- 5.32) in the GLA-supplemented group as compared with 8.28 +/- 6.54 in the placebo group (P = 0.09; P = 0.06 after adjustment for total serum IgE at baseline, age 1 wk), but no significant effects on the other atopic outcomes. The increase in GLA concentrations in plasma phospholipids between baseline and 3 mo was negatively associated with the severity of atopic dermatitis at 1 y (Spearman's correlation coefficient = -0.233, P = 0.013). There was no significant effect on total serum IgE concentration. CONCLUSION: Early supplementation with GLA in children at high familial risk does not prevent the expression of atopy as reflected by total serum IgE, but it tends to alleviate the severity of atopic dermatitis in later infancy in these children. Copyright © 2003 American Society for Clinical Nutrition [ABSTRACT FROM AUTHOR]

Published
2003
Full Text
View/download PDF

117. Clemmensen Reduction. X. The Synthesis and Acidolysis of Some Aryl Alkyl Substituted Cyclopropane-1,2-diols

Author

BR Davis, Mark G. Hinds, and Ppc Ting

Subjects
Green chemistry, chemistry.chemical_classification, Reaction mechanism, Chemistry, Aryl, Diol, General Chemistry, Clemmensen reduction, Medicinal chemistry, Cyclopropane, chemistry.chemical_compound, Biocatalysis, polycyclic compounds, Alkyl
Abstract

The synthesis of a number of 1-alkyl-2-arylcyclopropane-1,2-diols is reported. Reductive treatment of suitably substituted 1,3-diketones gave a cyclopropanediyl diacetate from which the diol was obtained. Reaction of these diols with acid gave α- hydroxy ketones , in which an alkyl group, rather than an aryl group, was adjacent to the carbonyl function. Clemmensen reduction of the 1,3-diketones gave products of rearrangement, generally similar to those obtained by acidolysis of the cyclopropane-1,2-diols. The relationship of these results to thermochemical studies on the stability of carboxonium ions is discussed.

Published
1992
Full Text
View/download PDF

118. Ceftriaxone-induced Liver Injury: A Case Report.

Author

PPC LOW, GH SIEW, and MY KIING

Subjects
*CEFTRIAXONE, *LUNG injuries, *MEDICAL care, *PATIENTS, *CONFERENCES & conventions
Abstract

Introduction: Ceftriaxone is a commonly used broad-spectrum third generation cephalosporin for systemic infection empirically especially in the Intensive Care Unit (ICU). The objective of this case report is to highlight the rare complication of ceftriaxone-induced liver injury to ensure early diagnosis and prompt cessation of ceftraixone in highly suspicious cases. Drug induced liver injury (DILI) is a disorder whereby administered drug causes liver damage and occurs shortly after exposure. In the past, there were a few cases of reported raised liver enzymes caused by ceftriaxone. Case description: This case study reports a 14-year-old girl with underlying Type 1 Diabetes Mellitus who was admitted for Diabetes Ketoacidosis (DKA) secondary to infective gastroenteritis. She was treated empirically with Ceftriaxone and developed an unexpected deranged liver function test. Her jaundiced appearance and hepatomegaly prompted a thorough investigation of the cause of the sudden acute liver injury. Upon ruling out infective and autoimmune causes, this further established the diagnosis of ceftriaxone-induced liver injury. Ceftriaxone was immediately witheld with high suspicion of the ceftriaxone-induced liver injury and she was started on a course of N-acetylcysteine (NAC). She was subsequently discharged home well and her liver function test normalised in 3 months' time. Conclusion: Early recognition of drug-induced liver injury is important and prompt cessation of the offending drug is recommended. [ABSTRACT FROM AUTHOR]

Published
2021

119. Long-term outcome in coma.

Author

Nayana PPC, Serane TV, Nalini P, Mahadevan S, Nayana, Prabha P, Serane, Tiroumourougane V, Nalini, P, and Mahadevan, S

Abstract

Objective: To assess the relationship between Modified Glasgow Coma Scale, its components, brain stem reflexes and long term functional outcome in children with acute non-traumatic coma.Method: MGCS and brainstem reflexes were assessed at 6 hourly intervals for 72 hours from the time of admission. The children were followed up regularly and functional outcome was assessed at the end of 9 months. The lowest score of the MGCS and worst brain stem reflexes were used for the analysis.Results: Higher total MGCS score and verbal response score had a significant positive correlation with better functional outcome as measured by GOS and intelligence quotient. There was no association between the language function and the initial MGCS, its components and brain stem reflexes. Lower verbal response (P = 0.005) was the only factor that was found to be individually associated with poorer long term GOS score and intelligence quotient by multivariate linear regression analysis.Conclusion: In the long term prediction of outcome in acute non-traumatic coma, MGCS is not useful. However, verbal response, a component of MGCS, correlates well with long term functional outcome and intelligence quotient. [ABSTRACT FROM AUTHOR]

Published
2005
Full Text
View/download PDF

121. [Hijos_ ¿Cuantos?

Author

Izquierdo, Francisco, il, Ibarra, imp, PPC, ed. lit, Gráficas Ibarra, imp, Echeverría, Lamberto de, 1919-1987, Izquierdo, Francisco, il, Ibarra, imp, PPC, ed. lit, Gráficas Ibarra, imp, and Echeverría, Lamberto de, 1919-1987

Abstract

Tít. tomado de cub, Mención de responsabilidad tomada del final del texto, Tít. tomado da capa, Mención de responsabilidade tomada da fin do texto

Published
1956

122. Nueva Semana Santa

Author

Izquierdo, Francisco, il, PPC, ed. lit, Gráficas Uguina, imp, Sánchez Aliseda, Casimiro, 1914-1960, Izquierdo, Francisco, il, PPC, ed. lit, Gráficas Uguina, imp, and Sánchez Aliseda, Casimiro, 1914-1960

Abstract

Mención de responsabilidade tomada de fin de texto

Published
1957

124. Understanding the public health value and defining preferred product characteristics for therapeutic human papillomavirus (HPV) vaccines : World Health Organization consultations, October 2021—March 2022

Author

Pontificia Universidad Javeriana. Facultad de Medicina. Departamento de Medicina Interna. Grupo de Oncología. Centro Javeriano de Oncología, Murillo, Raul, Prudden, Holly J., Achilles, Sharon L., Schocken, Celina, Broutet, Nathalie, Canfell, Karen, Akaba, Hiroki, Basu, Partha, Bhatla, Neerja, Chirenje, Z. Mike, Delany-Moretlwe, Sinead, Denny, Lynette, Gamage, Deepa G., Herrero, Rolando, Hutubessy, Raymond, Villa, Luisa Lina, Schiller, John T., Stanley, Margaret, Temmerman, Marleen, Zhao, Fanghui, Ogilvie, Gina, Kaslow, David C., Dull, Peter, Gottlieb, Sami L., Therapeutic HPV Vaccine PPC Expert Consultation Group, Pontificia Universidad Javeriana. Facultad de Medicina. Departamento de Medicina Interna. Grupo de Oncología. Centro Javeriano de Oncología, Murillo, Raul, Prudden, Holly J., Achilles, Sharon L., Schocken, Celina, Broutet, Nathalie, Canfell, Karen, Akaba, Hiroki, Basu, Partha, Bhatla, Neerja, Chirenje, Z. Mike, Delany-Moretlwe, Sinead, Denny, Lynette, Gamage, Deepa G., Herrero, Rolando, Hutubessy, Raymond, Villa, Luisa Lina, Schiller, John T., Stanley, Margaret, Temmerman, Marleen, Zhao, Fanghui, Ogilvie, Gina, Kaslow, David C., Dull, Peter, Gottlieb, Sami L., and Therapeutic HPV Vaccine PPC Expert Consultation Group

126. Key epidemiological indicators and spatial autocorrelation patterns across five waves of COVID-19 in Catalonia

Author

Belvis-Costes, Francesc Xavier, Aleta Casas, Alberto, Padilla-Pozo, Álvaro, Pericas, Juan M, Fernández-Gracia, Juan, Rodriguez, Jorge P, Institut Català de la Salut, [Belvis F] Research Group on Health Inequalities, Environment, and Employment Conditions (GREDS EMCONET), Department of Political and Social Sciences, Universitat Pompeu Fabra, Barcelona, Spain. Johns Hopkins University-Universitat Pompeu Fabra Public Policy Center (JHU-UPF PPC), Barcelona, Spain. [Aleta A] Institute for Biocomputation and Physics of Complex Systems (BIFI), University of Zaragoza, Zaragoza, Spain. [Padilla-Pozo Á] Research Group on Health Inequalities, Environment, and Employment Conditions (GREDS EMCONET), Department of Political and Social Sciences, Universitat Pompeu Fabra, Barcelona, Spain. Johns Hopkins University-Universitat Pompeu Fabra Public Policy Center (JHU-UPF PPC), Barcelona, Spain. Department of Sociology, Cornell University, Ithaca, New York, USA. [Pericàs JM] Research Group on Health Inequalities, Environment, and Employment Conditions (GREDS EMCONET), Department of Political and Social Sciences, Universitat Pompeu Fabra, Barcelona, Spain. Johns Hopkins University-Universitat Pompeu Fabra Public Policy Center (JHU-UPF PPC), Barcelona, Spain. Unitat d’Hepatologia, Servei de Medicina Interna, Vall d’Hebron Hospital Universitari, Barcelona, Spain. Vall d’Hebron Institut de Recerca (VHIR), Barcelona, Spain. CIBERehd, Barcelona, Spain. Infectious Disease Department, Hospital Clínic, Barcelona, Spain. [Fernández-Gracia J] Instituto de Física Interdisciplinar Y Sistemas Complejos IFISC (CSIC-UIB), Palma de Mallorca, Spain. [Rodríguez JP] Instituto de Física Interdisciplinar Y Sistemas Complejos IFISC (CSIC-UIB), Palma de Mallorca, Spain. Instituto Mediterráneo de Estudios Avanzados IMEDEA (CSIC-UIB), Esporles, Spain, and Vall d'Hebron Barcelona Hospital Campus

Subjects
COVID-19 (Malaltia) - Epidemiologia - Catalunya, Catalonia, Other subheadings::Other subheadings::/epidemiology [Other subheadings], Cataluña, virosis::infecciones por virus ARN::infecciones por Nidovirales::infecciones por Coronaviridae::infecciones por Coronavirus [ENFERMEDADES], Otros calificadores::Otros calificadores::/epidemiología [Otros calificadores], epidemiología y bioestadística::epidemiología::usos de la epidemiología::monitorización epidemiológica::brotes de enfermedades [SALUD PÚBLICA], Virus Diseases::RNA Virus Infections::Nidovirales Infections::Coronaviridae Infections::Coronavirus Infections [DISEASES], Epidemiology and Biostatistics::Epidemiology::Uses of Epidemiology::Epidemiological Monitoring::Disease Outbreaks [PUBLIC HEALTH]
Abstract

Epidemiology; Statistics Epidemiología; Estadísticas Epidemiologia; Estadístiques This research studies the evolution of COVID-19 crude incident rates, effective reproduction number R(t) and their relationship with incidence spatial autocorrelation patterns in the 19 months following the disease outbreak in Catalonia (Spain). A cross-sectional ecological panel design based on n = 371 health-care geographical units is used. Five general outbreaks are described, systematically preceded by generalized values of R(t) > 1 in the two previous weeks. No clear regularities concerning possible initial focus appear when comparing waves. As for autocorrelation, we identify a wave’s baseline pattern in which global Moran’s I increases rapidly in the first weeks of the outbreak to descend later. However, some waves significantly depart from the baseline. In the simulations, both baseline pattern and departures can be reproduced when measures aimed at reducing mobility and virus transmissibility are introduced. Spatial autocorrelation is inherently contingent on the outbreak phase and is also substantially modified by external interventions affecting human behavior. All the authors acknowledge funding from the Social Observatory of the “la Caixa” Foundation as part of the project LCF/PR/SR20/52550011. Partial funding was also received from the Spanish Ministry of Science and Innovation (PID2020-117029RB-I00). Joan Benach gratefully acknowledges the financial support by ICREA under the ICREA Academia programme. J. Fernández-Gracia was supported by Direcció General de Política Universitària i Recerca from the government of the Balearic Islands through the postdoctoral program Vicenç Mut. J. P. Rodríguez is supported by Juan de la Cierva Formacion program (Ref. FJC2019-040622-I) funded by MCIN/AEI/ https://doi.org/10.13039/501100011033.

Published
2023

129. Automated Assessment of Glottal Dysfunction Through Unified Acoustic Voice Analysis

Author

Yan Song, Ian McLoughlin, Hamid Sharifzadeh, Jacqui E. Allen, Olivier Perrotin, Singapore Institute of Technology [Singapore] (SIT), GIPSA - Cognitive Robotics, Interactive Systems, & Speech Processing (GIPSA-CRISSP), GIPSA Pôle Parole et Cognition (GIPSA-PPC), Grenoble Images Parole Signal Automatique (GIPSA-lab), Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP ), Université Grenoble Alpes (UGA)-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP ), Université Grenoble Alpes (UGA)-Grenoble Images Parole Signal Automatique (GIPSA-lab), Université Grenoble Alpes (UGA), School of Computing, Unitec Institute of Technology, Department of Otolaryngology, North Shore Hospital, and University of Science and Technology of China [Hefei] (USTC)

Subjects
Glottal flow model, Glottis, Speech production, Voice Quality, Computer science, Speech recognition, Laryngectomy, Acoustic voice analysis, Distorted speech, Speech Acoustics, 030507 speech-language pathology & audiology, 03 medical and health sciences, Speech and Hearing, 0302 clinical medicine, [INFO.INFO-TS]Computer Science [cs]/Signal and Image Processing, Discriminative model, Humans, 030223 otorhinolaryngology, Voice source, Source model, Class (computer programming), Acoustics, LPN and LVN, Whispers, Glottal flow, Otorhinolaryngology, Voice, 0305 other medical science, Speech reconstruction
Abstract

International audience; This paper uses the recent glottal flow model for iterative adaptive inverse filtering to analyze recordings from dysfunctional speakers, namely those with larynx-related impairment such as laryngectomy. The analytical model allows extraction of the voice source spectrum, described by a compact set of parameters. This single model is used to visualize and better understand speech production characteristics across impaired and nonimpaired voices. The analysis reveals some discriminative aspects of the source model which map to a physiological class description of those impairments. Furthermore, being based on analysis of source parameters only, it is complementary to any existing techniques of vocal-tract or phonetic analysis. The results indicate the potential for future automated speech reconstruction systems that adapt to the method of reconstruction required, as well as being useful for mainstream speech systems, such as ASR, in which front-end analysis can direct back-end models to suit characteristics of impaired speech.

Published
2022
Full Text
View/download PDF

130. Unsupervised speech enhancement with deep dynamical generative speech and noise models

Author

Lin, Xiaoyu, Leglaive, Simon, Girin, Laurent, Alameda-Pineda, Xavier, Vers des robots à l’intelligence sociale au travers de l’apprentissage, de la perception et de la commande (ROBOTLEARN), Inria Grenoble - Rhône-Alpes, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Université Grenoble Alpes (UGA), Institut d'Électronique et des Technologies du numéRique (IETR), Université de Rennes (UR)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-Nantes Université - pôle Sciences et technologie, Nantes Université (Nantes Univ)-Nantes Université (Nantes Univ), GIPSA - Cognitive Robotics, Interactive Systems, & Speech Processing (GIPSA-CRISSP), GIPSA Pôle Parole et Cognition (GIPSA-PPC), Grenoble Images Parole Signal Automatique (GIPSA-lab), Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP ), Université Grenoble Alpes (UGA)-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP ), Université Grenoble Alpes (UGA)-Grenoble Images Parole Signal Automatique (GIPSA-lab), Université Grenoble Alpes (UGA), ANR-19-P3IA-0003,MIAI,MIAI @ Grenoble Alpes(2019), ANR-19-CE33-0008,ML3RI,Apprentissage de bas-niveau d'ineractions robotiques multi-modales avec plusieurs personnes(2019), and European Project: 871245,H2020-EU.2.1.1. - INDUSTRIAL LEADERSHIP - Leadership in enabling and industrial technologies - Information and Communication Technologies (ICT),SPRING(2020)

Subjects
FOS: Computer and information sciences, Sound (cs.SD), Computer Science - Machine Learning, [INFO.INFO-TS]Computer Science [cs]/Signal and Image Processing, [INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG], Audio and Speech Processing (eess.AS), FOS: Electrical engineering, electronic engineering, information engineering, Computer Science - Sound, Machine Learning (cs.LG), Electrical Engineering and Systems Science - Audio and Speech Processing, [INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI]
Abstract

International audience; This work builds on a previous work on unsupervised speech enhancement using a dynamical variational autoencoder (DVAE) as the clean speech model and non-negative matrix factorization (NMF) as the noise model. We propose to replace the NMF noise model with a deep dynamical generative model (DDGM) depending either on the DVAE latent variables, or on the noisy observations, or on both. This DDGM can be trained in three configurations: noise-agnostic, noise-dependent and noise adaptation after noise-dependent training. Experimental results show that the proposed method achieves competitive performance compared to state-of-the-art unsupervised speech enhancement methods, while the noise-dependent training configuration yields a much more time-efficient inference process.

Published
2023

131. The Effect of Cued-Speech (CS) Perception on Auditory Processing in Typically Hearing (TH) Individuals Who Are Either Naïve or Experienced CS Producers

Author

Caron, Cora Jirschik, Vilain, Coriandre, Schwartz, Jean-Luc, Bayard, Clémence, Calcus, Axelle, Leybaert, Jacqueline, Colin, Cécile, Université libre de Bruxelles (ULB), GIPSA-Services (GIPSA-Services), Grenoble Images Parole Signal Automatique (GIPSA-lab), Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP ), Université Grenoble Alpes (UGA)-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP ), Université Grenoble Alpes (UGA), GIPSA - Perception, Contrôle, Multimodalité et Dynamiques de la parole (GIPSA-PCMD), GIPSA Pôle Parole et Cognition (GIPSA-PPC), Université Grenoble Alpes (UGA)-Grenoble Images Parole Signal Automatique (GIPSA-lab), and European Project: 860755,Marie Skłodowska-Curie Innovative Training Networks,Comm4CHILD(2020)

Subjects
cued speech, [SCCO.NEUR]Cognitive science/Neuroscience, event-related potentials, multimodality
Abstract

International audience; Cued Speech (CS) is a communication system that uses manual gestures to facilitate lipreading. In this study, we investigated how CS information interacts with natural speech using Event-Related Potential (ERP) analyses in French-speaking, typically hearing adults (TH) who were either naïve or experienced CS producers. The audiovisual (AV) presentation of lipreading information elicited an amplitude attenuation of the entire N1 and P2 complex in both groups, accompanied by N1 latency facilitation in the group of CS producers. Adding CS gestures to lipread information increased the magnitude of effects observed at the N1 time window, but did not enhance P2 amplitude attenuation. Interestingly, presenting CS gestures without lipreading information yielded distinct response patterns depending on participants’ experience with the system. In the group of CS producers, AV perception of CS gestures facilitated the early stage of speech processing, while in the group of naïve participants, it elicited a latency delay at the P2 time window. These results suggest that, for experienced CS users, the perception of gestures facilitates early stages of speech processing, but when people are not familiar with the system, the perception of gestures impacts the efficiency of phonological decoding.

Published
2023
Full Text
View/download PDF

132. A Multimodal Dynamical Variational Autoencoder for Audiovisual Speech Representation Learning

Author

Sadok, Samir, Leglaive, Simon, Girin, Laurent, Alameda-Pineda, Xavier, Séguier, Renaud, Institut d'Électronique et des Technologies du numéRique (IETR), Université de Rennes (UR)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-Nantes Université - pôle Sciences et technologie, Nantes Université (Nantes Univ)-Nantes Université (Nantes Univ), GIPSA - Cognitive Robotics, Interactive Systems, & Speech Processing (GIPSA-CRISSP), GIPSA Pôle Parole et Cognition (GIPSA-PPC), Grenoble Images Parole Signal Automatique (GIPSA-lab), Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP ), Université Grenoble Alpes (UGA)-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP ), Université Grenoble Alpes (UGA)-Grenoble Images Parole Signal Automatique (GIPSA-lab), Université Grenoble Alpes (UGA), Vers des robots à l’intelligence sociale au travers de l’apprentissage, de la perception et de la commande (ROBOTLEARN), Inria Grenoble - Rhône-Alpes, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Université Grenoble Alpes (UGA), ANR-19-P3IA-0003,MIAI,MIAI @ Grenoble Alpes(2019), ANR-19-CE33-0008,ML3RI,Apprentissage de bas-niveau d'ineractions robotiques multi-modales avec plusieurs personnes(2019), and European Project: 871245,H2020-EU.2.1.1. - INDUSTRIAL LEADERSHIP - Leadership in enabling and industrial technologies - Information and Communication Technologies (ICT),SPRING(2020)

Subjects
FOS: Computer and information sciences, Computer Science - Machine Learning, Sound (cs.SD), [INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG], Audio and Speech Processing (eess.AS), FOS: Electrical engineering, electronic engineering, information engineering, [INFO.INFO-CV]Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV], Computer Science - Sound, Computer Science - Multimedia, Electrical Engineering and Systems Science - Audio and Speech Processing, Machine Learning (cs.LG), Multimedia (cs.MM)
Abstract

In this paper, we present a multimodal \textit{and} dynamical VAE (MDVAE) applied to unsupervised audio-visual speech representation learning. The latent space is structured to dissociate the latent dynamical factors that are shared between the modalities from those that are specific to each modality. A static latent variable is also introduced to encode the information that is constant over time within an audiovisual speech sequence. The model is trained in an unsupervised manner on an audiovisual emotional speech dataset, in two stages. In the first stage, a vector quantized VAE (VQ-VAE) is learned independently for each modality, without temporal modeling. The second stage consists in learning the MDVAE model on the intermediate representation of the VQ-VAEs before quantization. The disentanglement between static versus dynamical and modality-specific versus modality-common information occurs during this second training stage. Extensive experiments are conducted to investigate how audiovisual speech latent factors are encoded in the latent space of MDVAE. These experiments include manipulating audiovisual speech, audiovisual facial image denoising, and audiovisual speech emotion recognition. The results show that MDVAE effectively combines the audio and visual information in its latent space. They also show that the learned static representation of audiovisual speech can be used for emotion recognition with few labeled data, and with better accuracy compared with unimodal baselines and a state-of-the-art supervised model based on an audiovisual transformer architecture., Comment: 25 pages, 14 figures, https://samsad35.github.io/site-mdvae/

Published
2023

133. Perception de parole chez les enfants porteurs d'implants cochléaires : Apports de l'Auditory Verbal Therapy et de la Langue française Parlée Complétée

Author

Van Bogaert, Lucie, Machart, Laura, Vilain, Anne, Loevenbruck, Hélène, Consortium, Eulalies, Laboratoire de Psychologie et NeuroCognition (LPNC ), Université Savoie Mont Blanc (USMB [Université de Savoie] [Université de Chambéry])-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA), GIPSA - Perception, Contrôle, Multimodalité et Dynamiques de la parole (GIPSA-PCMD), GIPSA Pôle Parole et Cognition (GIPSA-PPC), Grenoble Images Parole Signal Automatique (GIPSA-lab), Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP ), Université Grenoble Alpes (UGA)-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP ), Université Grenoble Alpes (UGA)-Grenoble Images Parole Signal Automatique (GIPSA-lab), Université Grenoble Alpes (UGA), ANR-19-CE28-0016,EULALIES,Un nouvel outil d'évaluation des Troubles du Développement des Sons de Parole chez les enfants francophones(2019), and European Project: 860755,Marie Skłodowska-Curie Innovative Training Networks,Comm4CHILD(2020)

Subjects
[SCCO]Cognitive science, [SCCO.LING]Cognitive science/Linguistics
Abstract

ISBN : 978-2-917490-35-8; International audience; Bien que les implants cochléaires (IC) améliorent la perception de la parole, les informations auditives transmises par celui-ci restent dégradées et imparfaites (Leybaert & LaSasso, 2010). Cette étude porte sur les bénéfices pour la perception de parole de deux approches de soutien de la langue vocale chez les enfants porteurs d’IC. L’Auditory Verbal Therapy (AVT) est un programme d'intervention précoce qui s'appuie principalement sur l’entrainement auditif pour que l’enfant sourd puisse développer des compétences auditives et langagières adéquates. La Langue française Parlée Complétée (LfPC) est un outil d’aide à la communication qui utilise la multisensorialité de la parole pour désambiguïser la lecture labiale en y ajoutant un geste manuel. Cette étude inclut des enfants âgés de 5 à 11 ans : 90 enfants normo-entendants (NE), 9 enfants porteurs d’IC ayant participé à un programme AVT (AVT), 6 enfants porteurs d’IC ayant un niveau élevé de décodage de la LfPC (LfPC+), et 19 enfants porteurs d’IC ayant un faible niveau de décodage de la LfPC (LfPC-). Le niveau de décodage été mesuré en utilisant le test TERMO (Busquet & Descourtieux, 2000). Dans cette étude, la tâche de jugement de lexicalité de la batterie EULALIES (Meloni et al. 2017) a été utilisée. Des mots étaient présentés en modalité audiovisuelle, avec altération phonologique ou non et l'enfant devait juger si ceux-ci étaient prononcés correctement. La sensibilité de la perception a été mesurée par un score d’ (Macmillan and Creelman, 2004). Un intervalle de prédiction à 95% a été calculé pour les enfants NE et les scores des enfants porteurs d’IC ont été comparés à cet intervalle (par des tests de conformité des proportions). Pour les scores IC en dehors de cet intervalle, la distance par rapport à la borne supérieure de l'intervalle a été calculée.Les résultats montrent que les enfants LfPC+ et LfPC- ont des performances significativement plus faibles que leurs pairs entendants et que les enfants AVT ont tendance à avoir des scores plus faibles. De plus, les scores des enfants des groupes AVT et LfPC+ sont plus proches de l’intervalle que ceux des enfants du groupe LfPC-. Ainsi, cette étude semble montrer que ces deux approches (AVT et LfPC) contribuent à la rééducation de la perception de parole, et souligne l'importance d'utiliser une approche spécifique en soutien de l’IC.Références bibliographiquesBUSQUET, D., & DESCOURTIEUX, C. (2000). TERMO Test d’évaluation de la réception du message oral par l’enfant sourd. Unithèque. https://www.unitheque.com/termo-test-evaluation-reception-message-oral-par-enfant-sourd/ortho/Livre/4074 LEYBAERT, J., LASASSO, C.J., Cued Speech for Enhancing Speech Perception and First Language Development of Children With Cochlear Implants. Trends Amplif, 2010, 96–112. MACMILLAN, N.A., CREELMAN, C.D., Detection Theory: A User’s Guide, 2nd ed. Psychology Press, New York, 2004.MELONI G., LOEVENBRUCK H., VILAIN A. & MACLEOD A. EULALIES, The France-Québec Speech Sound Disorders project, Actes de 14th International Congress for the Study of Child Language (IASCL), Lyon, France, 2017.

Published
2023

134. Speech Modeling with a Hierarchical Transformer Dynamical VAE

Author

Lin, Xiaoyu, Bie, Xiaoyu, Leglaive, Simon, Girin, Laurent, Alameda-Pineda, Xavier, Vers des robots à l’intelligence sociale au travers de l’apprentissage, de la perception et de la commande (ROBOTLEARN), Inria Grenoble - Rhône-Alpes, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Université Grenoble Alpes (UGA), Institut d'Électronique et des Technologies du numéRique (IETR), Université de Rennes (UR)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-Nantes Université - pôle Sciences et technologie, Nantes Université (Nantes Univ)-Nantes Université (Nantes Univ), GIPSA - Cognitive Robotics, Interactive Systems, & Speech Processing (GIPSA-CRISSP), GIPSA Pôle Parole et Cognition (GIPSA-PPC), Grenoble Images Parole Signal Automatique (GIPSA-lab), Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP ), Université Grenoble Alpes (UGA)-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP ), Université Grenoble Alpes (UGA)-Grenoble Images Parole Signal Automatique (GIPSA-lab), Université Grenoble Alpes (UGA), ANR-19-P3IA-0003,MIAI,MIAI @ Grenoble Alpes(2019), ANR-19-CE33-0008,ML3RI,Apprentissage de bas-niveau d'ineractions robotiques multi-modales avec plusieurs personnes(2019), and European Project: 871245,H2020-EU.2.1.1. - INDUSTRIAL LEADERSHIP - Leadership in enabling and industrial technologies - Information and Communication Technologies (ICT),SPRING(2020)

Subjects
FOS: Computer and information sciences, Sound (cs.SD), Computer Science - Machine Learning, [INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG], Audio and Speech Processing (eess.AS), [INFO.INFO-SD]Computer Science [cs]/Sound [cs.SD], FOS: Electrical engineering, electronic engineering, information engineering, Computer Science - Sound, Machine Learning (cs.LG), Electrical Engineering and Systems Science - Audio and Speech Processing, [INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI]
Abstract

International audience; The dynamical variational autoencoders (DVAEs) are a family of latent-variable deep generative models that extends the VAE to model a sequence of observed data and a corresponding sequence of latent vectors. In almost all the DVAEs of the literature, the temporal dependencies within each sequence and across the two sequences are modeled with recurrent neural networks. In this paper, we propose to model speech signals with the Hierarchical Transformer DVAE (HiT-DVAE), which is a DVAE with two levels of latent variable (sequence-wise and frame-wise) and in which the temporal dependencies are implemented with the Transformer architecture. We show that HiT-DVAE outperforms several other DVAEs for speech spectrogram modeling, while enabling a simpler training procedure, revealing its high potential for downstream low-level speech processing tasks such as speech enhancement.

Published
2023
Full Text
View/download PDF

135. Multivariate Analysis of RNA Chemistry Marks Uncovers Epitranscriptomics-Based Biomarker Signature for Adult Diffuse Glioma Diagnostics

Author

S. Relier, A. Amalric, A. Attina, I.B. Koumare, V. Rigau, F. Burel Vandenbos, D. Fontaine, M. Baroncini, J.P. Hugnot, H. Duffau, L. Bauchet, C. Hirtz, E. Rivals, A. David, Gouat, Isabelle, Institut de Génomique Fonctionnelle (IGF), Institut National de la Santé et de la Recherche Médicale (INSERM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM), Plateforme de Protéomique Clinique de Montpellier (PPC), BioCampus (BCM), Institut National de la Santé et de la Recherche Médicale (INSERM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Institut National de la Santé et de la Recherche Médicale (INSERM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM), Cellules Souches, Plasticité Cellulaire, Médecine Régénératrice et Immunothérapies (IRMB), Centre Hospitalier Régional Universitaire [Montpellier] (CHRU Montpellier)-Institut National de la Santé et de la Recherche Médicale (INSERM)-Université de Montpellier (UM), Service de Neurochirurgie [Montpellier], Centre Hospitalier Régional Universitaire [Montpellier] (CHRU Montpellier)-CHU Gui de Chauliac [Montpellier], Hôpital Gabriel Touré [Mali], Centre Hospitalier Régional Universitaire [Montpellier] (CHRU Montpellier), Centre Hospitalier Universitaire de Nice (CHU Nice), Centre Hospitalier Régional Universitaire [Lille] (CHRU Lille), Méthodes et Algorithmes pour la Bioinformatique (MAB), Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier (LIRMM), and Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)-Centre National de la Recherche Scientifique (CNRS)-Université de Montpellier (UM)

Subjects
Proteomics, [SDV.BIBS] Life Sciences [q-bio]/Quantitative Methods [q-bio.QM], Multivariate Analysis, Humans, Metabolomics, RNA, Glioma, [SDV.BIBS]Life Sciences [q-bio]/Quantitative Methods [q-bio.QM], Biomarkers, Analytical Chemistry
Abstract

International audience; One of the main challenges in cancer management relates to the discovery of reliable biomarkers, which could guide decision-making and predict treatment outcome. In particular, the rise and democratization of high-throughput molecular profiling technologies bolstered the discovery of "biomarker signatures" that could maximize the prediction performance. Such an approach was largely employed from diverse OMICs data (i.e., genomics, transcriptomics, proteomics, metabolomics) but not from epitranscriptomics, which encompasses more than 100 biochemical modifications driving the post-transcriptional fate of RNA: stability, splicing, storage, and translation. We and others have studied chemical marks in isolation and associated them with cancer evolution, adaptation, as well as the response to conventional therapy. In this study, we have designed a unique pipeline combining multiplex analysis of the epitranscriptomic landscape by high-performance liquid chromatography coupled to tandem mass spectrometry with statistical multivariate analysis and machine learning approaches in order to identify biomarker signatures that could guide precision medicine and improve disease diagnosis. We applied this approach to analyze a cohort of adult diffuse glioma patients and demonstrate the existence of an "epitranscriptomics-based signature" that permits glioma grades to be discriminated and predicted with unmet accuracy. This study demonstrates that epitranscriptomics (co)evolves along cancer progression and opens new prospects in the field of omics molecular profiling and personalized medicine.

Published
2022
Full Text
View/download PDF

136. Automatic assessment of oral readings of young pupils

Author

Gérard Bailly, Erika Godde, Anne-Laure Piat-Marchand, Marie-Line Bosse, GIPSA - Cognitive Robotics, Interactive Systems, & Speech Processing (GIPSA-CRISSP), GIPSA Pôle Parole et Cognition (GIPSA-PPC), Grenoble Images Parole Signal Automatique (GIPSA-lab), Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP ), Université Grenoble Alpes (UGA)-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP ), Université Grenoble Alpes (UGA)-Grenoble Images Parole Signal Automatique (GIPSA-lab), Université Grenoble Alpes (UGA), Laboratoire de Psychologie et NeuroCognition (LPNC ), Université Savoie Mont Blanc (USMB [Université de Savoie] [Université de Chambéry])-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA), CDC-founded e-FRAN Fluence, and ANR-19-P3IA-0003,MIAI,MIAI @ Grenoble Alpes(2019)

Subjects
Linguistics and Language, [SHS.EDU]Humanities and Social Sciences/Education, Communication, Modeling and Simulation, Computer Vision and Pattern Recognition, [SPI.SIGNAL]Engineering Sciences [physics]/Signal and Image processing, Language and Linguistics, Software, Computer Science Applications
Abstract

International audience; We propose a computational framework for estimating multidimensional subjective ratings of the reading performance of young readers from speech-based objective measures. We combine linguistic features (number of correct words, repetitions, deletions, insertions uttered per minute, etc.) with prosodic features. Expressivity is particularly difficult to predict since there is no unique gold standard. We propose a novel framework for performing such an estimation that exploits multiple references performed by adults and we demonstrate its effectiveness using recordings from a large data set of 1063 oral readings from 442 children (more than 30 h of speech), 84 oral readings from 42 adults and 6853 subjective scores delivered by 29 different human raters. We show that robust and accurate estimations of reading fluency can be achieved using combined features. This automatic assessment tool provides teachers and speech therapists with reliable estimates of the maturation of several reading skills.

Published
2022
Full Text
View/download PDF

137. Speech rehabilitation in children with cochlear implants using a multisensory (French Cued Speech) or a hearing-focused (Auditory Verbal Therapy) approach

Author

Van Bogaert, Lucie, Machart, Laura, Gerber, Silvain, Lœvenbruck, Hélène, Vilain, Anne, Consortium, Eulalies, Laboratoire de Psychologie et NeuroCognition (LPNC ), Université Savoie Mont Blanc (USMB [Université de Savoie] [Université de Chambéry])-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA), GIPSA - Perception, Contrôle, Multimodalité et Dynamiques de la parole (GIPSA-PCMD), GIPSA Pôle Parole et Cognition (GIPSA-PPC), Grenoble Images Parole Signal Automatique (GIPSA-lab), Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP ), Université Grenoble Alpes (UGA)-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP ), Université Grenoble Alpes (UGA)-Grenoble Images Parole Signal Automatique (GIPSA-lab), Université Grenoble Alpes (UGA), GIPSA-Services (GIPSA-Services), ANR-19-CE28-0016,EULALIES,Un nouvel outil d'évaluation des Troubles du Développement des Sons de Parole chez les enfants francophones(2019), and European Project: 860755,Marie Skłodowska-Curie Innovative Training Networks,Comm4CHILD(2020)

Subjects
[SCCO]Cognitive science, speech rehabilitation, children, French Cued Speech, cochlear implant, hearing impairment, [SHS.LANGUE]Humanities and Social Sciences/Linguistics, speech perception, Auditory Verbal Therapy, [SHS]Humanities and Social Sciences
Abstract

International audience; Introduction Early exposure to a rich linguistic environment is essential as soon as the diagnosis of deafness is made. Cochlear implantation (CI) allows children to have access to speech perception in their early years. However, it provides only partial acoustic information, which can lead to difficulties in perceiving some phonetic contrasts. This study investigates the contribution of two spoken speech and language rehabilitation approaches to speech perception in children with CI using a lexicality judgment task from the EULALIES battery. Auditory Verbal Therapy (AVT) is an early intervention program that relies on auditory learning to enhance hearing skills in deaf children with CI. French Cued Speech, also called Cued French (CF), is a multisensory communication tool that disambiguates lip reading by adding a manual gesture. Methods In this study, 124 children aged from 60 to 140 months were included: 90 children with typical hearing skills (TH), 9 deaf children with CI who had participated in an AVT program (AVT), 6 deaf children with CI with high Cued French reading skills (CF+), and 19 deaf children with CI with low Cued French reading skills (CF-). Speech perception was assessed using sensitivity ( d’ ) using both the hit and false alarm rates, as defined in signal-detection theory. Results The results show that children with cochlear implants from the CF- and CF+ groups have significantly lower performance compared to children with typical hearing (TH) ( p < 0.001 and p = 0.033, respectively). Additionally, children in the AVT group also tended to have lower scores compared to TH children ( p = 0.07). However, exposition to AVT and CF seems to improve speech perception. The scores of the children in the AVT and CF+ groups are closer to typical scores than those of children in the CF- group, as evidenced by a distance measure. Discussion Overall, the findings of this study provide evidence for the effectiveness of these two speech and language rehabilitation approaches, and highlight the importance of using a specific approach in addition to a cochlear implant to improve speech perception in children with cochlear implants.

Published
2023
Full Text
View/download PDF

138. A Multistream Model for Continuous Recognition of Lexical Units in French Sign Language

Author

Ouakrim, Yanis, Beautemps, Denis, Gouiffès, Michèle, Hueber, Thomas, Berthommier, Frédéric, Braffort, Annelies, GIPSA - Cognitive Robotics, Interactive Systems, & Speech Processing (GIPSA-CRISSP), GIPSA Pôle Parole et Cognition (GIPSA-PPC), Grenoble Images Parole Signal Automatique (GIPSA-lab), Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP ), Université Grenoble Alpes (UGA)-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP ), Université Grenoble Alpes (UGA)-Grenoble Images Parole Signal Automatique (GIPSA-lab), Université Grenoble Alpes (UGA), Laboratoire Interdisciplinaire des Sciences du Numérique (LISN), Institut National de Recherche en Informatique et en Automatique (Inria)-CentraleSupélec-Université Paris-Saclay-Centre National de la Recherche Scientifique (CNRS), and BPI France, projet serveur gestuel

Subjects
[INFO.INFO-TS]Computer Science [cs]/Signal and Image Processing, [INFO.INFO-CV]Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV], [INFO.INFO-CL]Computer Science [cs]/Computation and Language [cs.CL]
Abstract

Sign languages are the primary means of communication for deaf people, yet they remain poorly endowed from a natural language processing perspective, thus linguistic tools (translators, concordancers) dedicated to them are rare. In this paper, starting from a corpus limited size, we propose a very first learning model for the recognition of lexical signs in French Sign Language (LSF). Signers are represented by sequences of 3D pose, which are decoded into sign sequences thanks to a bidirectionnal recurrent neural network trained with a CTC loss (Connectionist Temporal Classification). Different strategies for combining the different articulators of LSF within the neural decoder (single vs. multistream architecture), as well as the contribution of main articulators to the overall performance of the system, are investigated.; Les langues des signes constituent le premier moyen de communication des personnes sourdes, pourtant elles restent peu dotées du point de vue du traitement automatique des langues, et les outils linguistiques (traducteurs, concordanciers) qui leur sont dédiés sont rares. Dans ce papier, partant d'un corpus de dialogues en LSF de taille limitée, nous proposons un tout premier modèle d'apprentissage pour la reconnaissance de signes lexicaux en Langue des Signes Française (LSF). Les signeurs sont représentés par leur pose 3D, qui est fournie à un réseau de neurone récurrent bi-directionnel, entrainé à l'aide de l'approche CTC (Connectionist Temporal Classification). L'apport des principaux articulateurs est évalué dans la reconnaissance de la LSF et des approches mono et multiflux (un modèle par articulateur) sont comparées.

Published
2023

139. Learning and controlling the source-filter representation of speech with a variational autoencoder

Author

Samir Sadok, Simon Leglaive, Laurent Girin, Xavier Alameda-Pineda, Renaud Séguier, Institut d'Électronique et des Technologies du numéRique (IETR), Université de Rennes (UR)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-Nantes Université - pôle Sciences et technologie, Nantes Université (Nantes Univ)-Nantes Université (Nantes Univ), CentraleSupélec [campus de Rennes], GIPSA - Cognitive Robotics, Interactive Systems, & Speech Processing (GIPSA-CRISSP), GIPSA Pôle Parole et Cognition (GIPSA-PPC), Grenoble Images Parole Signal Automatique (GIPSA-lab), Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP ), Université Grenoble Alpes (UGA)-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP ), Université Grenoble Alpes (UGA)-Grenoble Images Parole Signal Automatique (GIPSA-lab), Université Grenoble Alpes (UGA), Vers des robots à l’intelligence sociale au travers de l’apprentissage, de la perception et de la commande (ROBOTLEARN), Inria Grenoble - Rhône-Alpes, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Université Grenoble Alpes (UGA), ANR-19-P3IA-0003,MIAI,MIAI @ Grenoble Alpes(2019), ANR-19-CE33-0008,ML3RI,Apprentissage de bas-niveau d'ineractions robotiques multi-modales avec plusieurs personnes(2019), European Project: 871245,H2020-EU.2.1.1. - INDUSTRIAL LEADERSHIP - Leadership in enabling and industrial technologies - Information and Communication Technologies (ICT),SPRING(2020), Société Française d'Acoustique (SFA), Université de Rennes 1 (UR1), Université de Rennes (UNIV-RENNES)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Université de Rennes (UNIV-RENNES)-Institut National des Sciences Appliquées (INSA)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-Nantes Université - pôle Sciences et technologie, and Leglaive, Simon

Subjects
[INFO.INFO-AI] Computer Science [cs]/Artificial Intelligence [cs.AI], FOS: Computer and information sciences, Computer Science - Machine Learning, Sound (cs.SD), Linguistics and Language, Deep generative models, Variational autoencoder, Computer Science - Sound, Language and Linguistics, Representation learning, Machine Learning (cs.LG), [INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI], [INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG], Audio and Speech Processing (eess.AS), FOS: Electrical engineering, electronic engineering, information engineering, [SPI.SIGNAL] Engineering Sciences [physics]/Signal and Image processing, Communication, [INFO.INFO-LG] Computer Science [cs]/Machine Learning [cs.LG], [INFO.INFO-SD] Computer Science [cs]/Sound [cs.SD], Computer Science Applications, Source-filter model, Modeling and Simulation, [INFO.INFO-SD]Computer Science [cs]/Sound [cs.SD], Computer Vision and Pattern Recognition, [SPI.SIGNAL]Engineering Sciences [physics]/Signal and Image processing, Software, Electrical Engineering and Systems Science - Audio and Speech Processing
Abstract

Understanding and controlling latent representations in deep generative models is a challenging yet important problem for analyzing, transforming and generating various types of data. In speech processing, inspiring from the anatomical mechanisms of phonation, the source-filter model considers that speech signals are produced from a few independent and physically meaningful continuous latent factors, among which the fundamental frequency $f_0$ and the formants are of primary importance. In this work, we start from a variational autoencoder (VAE) trained in an unsupervised manner on a large dataset of unlabeled natural speech signals, and we show that the source-filter model of speech production naturally arises as orthogonal subspaces of the VAE latent space. Using only a few seconds of labeled speech signals generated with an artificial speech synthesizer, we propose a method to identify the latent subspaces encoding $f_0$ and the first three formant frequencies, we show that these subspaces are orthogonal, and based on this orthogonality, we develop a method to accurately and independently control the source-filter speech factors within the latent subspaces. Without requiring additional information such as text or human-labeled data, this results in a deep generative model of speech spectrograms that is conditioned on $f_0$ and the formant frequencies, and which is applied to the transformation speech signals. Finally, we also propose a robust $f_0$ estimation method that exploits the projection of a speech signal onto the learned latent subspace associated with $f_0$., 23 pages, 7 figures, companion website: https://samsad35.github.io/site-sfvae/

Published
2023
Full Text
View/download PDF

140. Unsupervised Speech Enhancement Using Dynamical Variational Autoencoders

Author

Xiaoyu Bie, Simon Leglaive, Xavier Alameda-Pineda, Laurent Girin, Vers des robots à l’intelligence sociale au travers de l’apprentissage, de la perception et de la commande (ROBOTLEARN), Inria Grenoble - Rhône-Alpes, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Université Grenoble Alpes (UGA), CentraleSupélec, Institut d'Électronique et des Technologies du numéRique (IETR), Université de Rennes (UR)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-Nantes Université - pôle Sciences et technologie, Nantes Université (Nantes Univ)-Nantes Université (Nantes Univ), GIPSA - Cognitive Robotics, Interactive Systems, & Speech Processing (GIPSA-CRISSP), GIPSA Pôle Parole et Cognition (GIPSA-PPC), Grenoble Images Parole Signal Automatique (GIPSA-lab), Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP ), Université Grenoble Alpes (UGA)-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP ), Université Grenoble Alpes (UGA)-Grenoble Images Parole Signal Automatique (GIPSA-lab), Université Grenoble Alpes (UGA), ANR-19-P3IA-0003, ANR-3IA MIAI, ANR-19-CE33-0008-01, ANR-JCJC ML3RI, GA #871245, EC, ANR-19-P3IA-0003,MIAI,MIAI @ Grenoble Alpes(2019), ANR-19-CE33-0008,ML3RI,Apprentissage de bas-niveau d'ineractions robotiques multi-modales avec plusieurs personnes(2019), European Project: 871245,H2020-EU.2.1.1. - INDUSTRIAL LEADERSHIP - Leadership in enabling and industrial technologies - Information and Communication Technologies (ICT),SPRING(2020), and European Project: H2020,SPRING

Subjects
Computer Science::Machine Learning, FOS: Computer and information sciences, Computer Science - Machine Learning, Sound (cs.SD), Acoustics and Ultrasonics, Computer Science - Artificial Intelligence, Noise measurement, Speech enhancement, Time series analysis, Computer Science - Sound, [INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI], Machine Learning (cs.LG), Time-domain analysis, Computational Mathematics, Artificial Intelligence (cs.AI), [INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG], Computer Science::Sound, Audio and Speech Processing (eess.AS), Recording, FOS: Electrical engineering, electronic engineering, information engineering, Computer Science (miscellaneous), Training, Inference algorithms, Electrical and Electronic Engineering, Electrical Engineering and Systems Science - Audio and Speech Processing
Abstract

International audience; Dynamical variational autoencoders (DVAEs) are a class of deep generative models with latent variables, dedicated to model time series of high-dimensional data. DVAEs can be considered as extensions of the variational autoencoder (VAE) that include temporal dependencies between successive observed and/or latent vectors. Previous work has shown the interest of using DVAEs over the VAE for speech spectrograms modeling. Independently, the VAE has been successfully applied to speech enhancement in noise, in an unsupervised noise-agnostic set-up that requires neither noise samples nor noisy speech samples at training time, but only requires clean speech signals. In this paper, we extend these works to DVAE-based single-channel unsupervised speech enhancement, hence exploiting both speech signals unsupervised representation learning and dynamics modeling. We propose an unsupervised speech enhancement algorithm that combines a DVAE speech prior pre-trained on clean speech signals with a noise model based on nonnegative matrix factorization, and we derive a variational expectation-maximization (VEM) algorithm to perform speech enhancement. The algorithm is presented with the most general DVAE formulation and is then applied with three specific DVAE models to illustrate the versatility of the framework. Experimental results show that the proposed DVAE-based approach outperforms its VAE-based counterpart, as well as several supervised and unsupervised noise-dependent baselines, especially when the noise type is unseen during training.

Published
2022
Full Text
View/download PDF

141. Inflammatory Responses Induced by the Monophasic Variant of Salmonella Typhimurium in Pigs Play a Role in the High Shedder Phenotype and Fecal Microbiota Composition

Author

Kempf, Florent, Cordoni, Guido, Chaussé, Anne-Marie, Drumo, Rosanna, Brown, Helen, Horton, Daniel L., Paboeuf, Frédéric, Denis, Martine, Velge, Philippe, La Ragione, Roberto, Kerouanton, Annaëlle, Infectiologie et Santé Publique (UMR ISP), Université de Tours (UT)-Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement (INRAE), University of Surrey (UNIS), Laboratoire de Ploufragan-Plouzané-Niort [ANSES], Agence nationale de sécurité sanitaire de l'alimentation, de l'environnement et du travail (ANSES), School of Biosciences and Medicine [Surrey, UK], and European Project: 773830,H2020-SFS-2017-1 ,MoMIR-PPC (a component of European Joint Programme One Health) (2018)

Subjects
pig, [SDV.BA.MVSA]Life Sciences [q-bio]/Animal biology/Veterinary medicine and animal Health, gut microbiota, Physiology, high shedder, immunity, [SDV.MP.BAC]Life Sciences [q-bio]/Microbiology and Parasitology/Bacteriology, Biochemistry, Microbiology, Computer Science Applications, Salmonella, inflammation, Modeling and Simulation, Genetics, Molecular Biology, Ecology, Evolution, Behavior and Systematics
Abstract

International audience; Pigs infected with Salmonella may excrete large amounts of Salmonella, increasing the risk of spread of this pathogen in the food chain. Identifying Salmonella high shedder pigs is therefore required to mitigate this risk. We analyzed immune-associated markers and composition of the gut microbiota in specific-pathogen-free pigs presenting different shedding levels after an oral infection with Salmonella. Immune response was studied through total blood cell counts, production of anti-Salmonella antibodies and cytokines, and gene expression quantification. Total Salmonella shedding for each pig was estimated and hierarchical clustering was used to cluster pigs into high, intermediate, and low shedders. Gut microbiota compositions were assessed using 16S rRNA microbial community profiling. Comparisons were made between control and inoculated pigs, then between high and low shedders pigs. Prior to infection, high shedders had similar immunological profiles compared to low shedders. As soon as 1 day postinoculation (dpi), significant differences on the cytokine production level and on the expression level of several host genes related to a proinflammatory response were observed between high and low shedders. Infection with Salmonella induced an early and profound remodeling of the immune response in all pigs, but the intensity of the response was stronger in high shedders. In contrast, low shedders seroconverted earlier than high shedders. Just after induction of the proinflammatory response (at 2 dpi), some taxa of the fecal microbiota were specific to the shedding phenotypes. This was related to the enrichment of several functional pathways related to anaerobic respiration in high shedders. In conclusion, our data show that the immune response to Salmonella modifies the fecal microbiota and subsequently could be responsible for shedding phenotypes. Influencing the gut microbiota and reducing intestinal inflammation could be a strategy for preventing Salmonella high shedding in livestock. IMPORTANCE Salmonellosis remains the most frequent human foodborne zoonosis after campylobacteriosis and pork meat is considered one of the major sources of human foodborne infections. At the farm, host heterogeneity in pig infection is problematic. High Salmonella shedders contribute more significantly to the spread of this foodborne pathogen in the food chain. The identification of predictive biomarkers for high shedders could help to control Salmonella in pigs. The purpose of the present study was to investigate why some pigs become super shedders and others low shedders. We thus investigated the differences in the fecal microbial composition and the immune response in orally infected pigs presenting different Salmonella shedding patterns. Our data show that the proinflammatory response induced by S. Typhimurium at 1 dpi could be responsible for the modification of the fecal microbiota composition and functions observed mainly at 2 and 3 dpi and to the low and super shedder phenotypes.

Published
2023
Full Text
View/download PDF

142. Prosodie du corse

Author

Romano, Antonio, Boula De Mareüil, Philippe, Lai, Jean-Pierre, Laboratorio di Fonetica Sperimentale 'Arturo Genre' (LFSAG), Università degli studi di Torino (UNITO), Laboratoire d'Informatique pour la Mécanique et les Sciences de l'Ingénieur (LIMSI), Université Paris-Saclay-Centre National de la Recherche Scientifique (CNRS), GIPSA-Services (GIPSA-Services), Grenoble Images Parole Signal Automatique (GIPSA-lab), Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP ), Université Grenoble Alpes (UGA)-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP ), Université Grenoble Alpes (UGA), GIPSA - SYstèmes Linguistiques, Dialectologie et Oralité (GIPSA-SYLDO), GIPSA Pôle Parole et Cognition (GIPSA-PPC), Université Grenoble Alpes (UGA)-Grenoble Images Parole Signal Automatique (GIPSA-lab), Università degli studi di Torino = University of Turin (UNITO), Traitement du Langage Parlé (TLP ), Laboratoire Interdisciplinaire des Sciences du Numérique (LISN), Institut National de Recherche en Informatique et en Automatique (Inria)-CentraleSupélec-Université Paris-Saclay-Centre National de la Recherche Scientifique (CNRS)-Institut National de Recherche en Informatique et en Automatique (Inria)-CentraleSupélec-Université Paris-Saclay-Centre National de la Recherche Scientifique (CNRS)-Sciences et Technologies des Langues (STL), and Institut National de Recherche en Informatique et en Automatique (Inria)-CentraleSupélec-Université Paris-Saclay-Centre National de la Recherche Scientifique (CNRS)-Institut National de Recherche en Informatique et en Automatique (Inria)-CentraleSupélec-Université Paris-Saclay-Centre National de la Recherche Scientifique (CNRS)

Subjects
ComputingMilieux_MISCELLANEOUS, [SHS]Humanities and Social Sciences
Abstract

International audience

Published
2023

143. Correlation between the Effect of Orofacial Somatosensory Inputs in Speech Perception and Speech Production Performance

Author

Monica Ashokumar, Clément Guichet, Jean-Luc Schwartz, Takayuki Ito, GIPSA - Perception, Contrôle, Multimodalité et Dynamiques de la parole (GIPSA-PCMD), GIPSA Pôle Parole et Cognition (GIPSA-PPC), Grenoble Images Parole Signal Automatique (GIPSA-lab), Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP ), Université Grenoble Alpes (UGA)-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP ), Université Grenoble Alpes (UGA)-Grenoble Images Parole Signal Automatique (GIPSA-lab), Université Grenoble Alpes (UGA), Haskins Laboratories, This work was supported by the European Union’s Horizon 2020 research and innovation programme under the Marie Sklowdoska-Curie Grant Agreement No 860755 (Comm4CHILD project) and by National Institute on Deafness and Other Communication Disorders Grant R01-DC017439., and European Project: 860755,Marie Skłodowska-Curie Innovative Training Networks,Comm4CHILD(2020)

Subjects
[SCCO]Cognitive science, production-perception link, vowel categorization, Applied Mathematics, somatosensory feedback, Auditory-somatosensory integration
Abstract

International audience; Orofacial somatosensory inputs modify the perception of speech sounds. Such auditory-somatosensory integration likely develops alongside speech production acquisition. We examined whether the somatosensory effect in speech perception varies depending on individual characteristics of speech production. The somatosensory effect in speech perception was assessed by changes in category boundary between /e/ and /ø/ in a vowel identification test resulting from somatosensory stimulation providing facial skin deformation in the rearward direction corresponding to articulatory movement for /e/ applied together with the auditory input. Speech production performance was quantified by the acoustic distances between the average first, second and third formants of /e/ and /ø/ utterances recorded in a separate test. The category boundary between /e/ and /ø/ was significantly shifted toward /ø/ due to the somatosensory stimulation which is consistent with previous research. The amplitude of the category boundary shift was significantly correlated with the acoustic distance between the mean second – and marginally third – formants of /e/ and /ø/ productions, with no correlation with the first formant distance. Greater acoustic distances can be related to larger contrasts between the articulatory targets of vowels in speech production. These results suggest that the somatosensory effect in speech perception can be linked to speech production performance.

Published
2023
Full Text
View/download PDF

144. Désignation et représentation des éléments topographiques dans les dialectes de France : synthèse générale (relief, cours d’eau, cavernes)

Author

Philippe del Giudice, GIPSA - SYstèmes Linguistiques, Dialectologie et Oralité (GIPSA-SYLDO), GIPSA Pôle Parole et Cognition (GIPSA-PPC), Grenoble Images Parole Signal Automatique (GIPSA-lab), Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP ), Université Grenoble Alpes (UGA)-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP ), Université Grenoble Alpes (UGA)-Grenoble Images Parole Signal Automatique (GIPSA-lab), Université Grenoble Alpes (UGA), and ANR-15-CE38-0002,ECLATS,Extraction automatisée des Contenus géoLinguistiques d'ATlas et analyse Spatiale: application à la Dialectologie(2015)

Subjects
Motivation, Linguistics and Language, Cours d'eau, Sémantique, Montagne, [SHS.LANGUE]Humanities and Social Sciences/Linguistics, Cavernes, Dialecte, Language and Linguistics
Abstract

As the second part of a general study about semantic determinism, this article continues to analyze the topographic lexicon of Gallo-Romance dialects. The new concepts (‘river’, ‘brook’, ‘valley’, ‘cave’) that I examine in this paper complete my previous survey about the designations of hills and mountains. Most of all, the new set of data allows me to go beyond isolated concepts and to present for the first time the motivational synthesis of a whole theme. The result is that, whatever the concept, words referring to topography are generally created according to four matrices of designation: such words originally allude to (1) level; (2) concavity/convexity; (3) physical composition; or stem from a (4) conceptual confusion due to adjacency. The method that reduces hundreds of lexical forms to a handful of creative patterns leads to a direct perception of how the lexicon is structured and has a strong heuristic potential.

Published
2021
Full Text
View/download PDF

145. A multi-scale epidemic model of enteric infection with heterogeneous shedding resulting from host-microbiota-pathogen interactions

Author

Labarthe, Simon, Laroche, Béatrice, Polizzi, Bastien, Patout, Florian, Ribot, Magali, Mathématiques et Informatique Appliquées du Génome à l'Environnement [Jouy-En-Josas] (MaIAGE), Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement (INRAE), Biodiversité, Gènes & Communautés (BioGeCo), Université de Bordeaux (UB)-Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement (INRAE), from patterns to models in computational biodiversity and biotechnology (PLEIADE), Laboratoire Bordelais de Recherche en Informatique (LaBRI), Université de Bordeaux (UB)-École Nationale Supérieure d'Électronique, Informatique et Radiocommunications de Bordeaux (ENSEIRB)-Centre National de la Recherche Scientifique (CNRS)-Université de Bordeaux (UB)-École Nationale Supérieure d'Électronique, Informatique et Radiocommunications de Bordeaux (ENSEIRB)-Centre National de la Recherche Scientifique (CNRS)-Inria Bordeaux - Sud-Ouest, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Biodiversité, Gènes & Communautés (BioGeCo), Université de Bordeaux (UB)-Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement (INRAE)-Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement (INRAE), Dynamiques de populations multi-échelles pour des systèmes physiologiques (MUSCA), Inria Saclay - Ile de France, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Physiologie de la reproduction et des comportements [Nouzilly] (PRC), Institut Français du Cheval et de l'Equitation [Saumur] (IFCE)-Université de Tours (UT)-Centre National de la Recherche Scientifique (CNRS)-Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement (INRAE)-Institut Français du Cheval et de l'Equitation [Saumur] (IFCE)-Université de Tours (UT)-Centre National de la Recherche Scientifique (CNRS)-Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement (INRAE)-Mathématiques et Informatique Appliquées du Génome à l'Environnement [Jouy-En-Josas] (MaIAGE), Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement (INRAE)-Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement (INRAE), Laboratoire de Mathématiques de Besançon (UMR 6623) (LMB), Centre National de la Recherche Scientifique (CNRS)-Université de Franche-Comté (UFC), Université Bourgogne Franche-Comté [COMUE] (UBFC)-Université Bourgogne Franche-Comté [COMUE] (UBFC), Biostatistique et Processus Spatiaux (BioSP), Institut Denis Poisson (IDP), Université d'Orléans (UO)-Université de Tours (UT)-Centre National de la Recherche Scientifique (CNRS), and European Project: 773830,H2020-SFS-2017-1 ,MoMIR-PPC (a component of European Joint Programme One Health) (2018)

Subjects
[MATH.MATH-AP]Mathematics [math]/Analysis of PDEs [math.AP], [INFO.INFO-MO]Computer Science [cs]/Modeling and Simulation
Abstract

International audience; Enteric pathogens colonize the digestive tract of farm livestock, such as chickens or pigs and potentially infect food products, which results in a threat for human health and important economic losses. It has been shown that the ability to excrete the pathogen in the environment and contaminate other animals is variable. This heterogeneity in pathogen carriage and shedding results from interactions between the host’s immune response, the pathogen and the commensal intestinal microbiota and plays and important role in epidemic propagation and control. The objective here is to propose a theoretical model of pathogen excretion and transmission linking within and between host dynamic.

Published
2022

146. Language familiarity influences own‐race face recognition in 9‐ and 12‐month‐old infants

Author

Olivier Pascalis, David Méary, Anne Hillairet de Boisferon, Mathilde Fort, Anne Vilain, Gudrun Schwarzer, Judit Gervain, Claudia Kubicek, Hélène Loevenbruck, Laboratoire de Psychologie et NeuroCognition (LPNC ), Université Savoie Mont Blanc (USMB [Université de Savoie] [Université de Chambéry])-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA), Justus-Liebig-University [Gießen, Germany], Université Sorbonne Paris Cité (USPC), GIPSA - Perception, Contrôle, Multimodalité et Dynamiques de la parole (GIPSA-PCMD), GIPSA Pôle Parole et Cognition (GIPSA-PPC), Grenoble Images Parole Signal Automatique (GIPSA-lab), Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP ), Université Grenoble Alpes (UGA)-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP ), Université Grenoble Alpes (UGA)-Grenoble Images Parole Signal Automatique (GIPSA-lab), Université Grenoble Alpes (UGA), Institut des sciences cognitives Marc Jeannerod - Centre de neuroscience cognitive - UMR5229 (CNC), Université Claude Bernard Lyon 1 (UCBL), and Université de Lyon-Université de Lyon-Centre National de la Recherche Scientifique (CNRS)

Subjects
Male, First language, Face (sociological concept), Facial recognition system, 050105 experimental psychology, German, non-native language, Race (biology), Child Development, Developmental and Educational Psychology, Humans, native language, 0501 psychology and cognitive sciences, Nursery rhyme, 10. No inequality, ComputingMilieux_MISCELLANEOUS, Language, [SDV.NEU.PC]Life Sciences [q-bio]/Neurons and Cognition [q-bio.NC]/Psychology and behavior, 05 social sciences, face, Infant, Recognition, Psychology, language.human_language, Linguistics, Pediatrics, Perinatology and Child Health, language, Female, Test phase, recognition, Psychology, Facial Recognition, 050104 developmental & child psychology
Abstract

During their first year, infants attune to the faces and language(s) that are frequent in their environment. The present study investigates the impact of language familiarity on how French-learning 9- and 12-month-olds recognize own-race faces. In Experiment 1, infants were familiarized with the talking face of a Caucasian bilingual German-French speaker reciting a nursery rhyme in French (native condition) or in German (non-native condition). In the test phase, infants' face recognition was tested by presenting a picture of the speaker's face they were familiarized with, side by side with a novel face. At 9 and 12 months, neither infants in the native condition nor the ones in the non-native condition clearly recognized the speaker's face. In Experiment 2, we familiarized infants with the still picture of the speaker's face, along with the auditory speech stream. This time, both 9- and 12-month-olds recognized the face of the speaker they had been familiarized with, but only if she spoke in their native language. This study shows that at least from 9 months of age, language modulates the way faces are recognized.

Published
2021
Full Text
View/download PDF

147. A survey of sound source localization with deep learning methods

Author

Pierre-Amaury Grumiaux, Srđan Kitić, Laurent Girin, Alexandre Guérin, Laboratoire des Sciences du Numérique de Nantes (LS2N), Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-IMT Atlantique (IMT Atlantique), Institut Mines-Télécom [Paris] (IMT)-Institut Mines-Télécom [Paris] (IMT)-École Centrale de Nantes (Nantes Univ - ECN), Nantes Université (Nantes Univ)-Nantes Université (Nantes Univ)-Nantes université - UFR des Sciences et des Techniques (Nantes univ - UFR ST), Nantes Université - pôle Sciences et technologie, Nantes Université (Nantes Univ)-Nantes Université (Nantes Univ)-Nantes Université - pôle Sciences et technologie, Nantes Université (Nantes Univ), Orange Labs [Cesson-Sévigné], Orange Labs, GIPSA - Cognitive Robotics, Interactive Systems, & Speech Processing (GIPSA-CRISSP), GIPSA Pôle Parole et Cognition (GIPSA-PPC), Grenoble Images Parole Signal Automatique (GIPSA-lab), Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP ), Université Grenoble Alpes (UGA)-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP ), Université Grenoble Alpes (UGA)-Grenoble Images Parole Signal Automatique (GIPSA-lab), Université Grenoble Alpes (UGA), ANRT CIFRE 2019/0533, ANR-19-P3IA-0003, and ANR-19-P3IA-0003,MIAI,MIAI @ Grenoble Alpes(2019)

Subjects
FOS: Computer and information sciences, [SPI.ACOU]Engineering Sciences [physics]/Acoustics [physics.class-ph], Sound (cs.SD), Computer Science - Machine Learning, Acoustics and Ultrasonics, deep learning, audio processing, Computer Science - Sound, Machine Learning (cs.LG), sound source localization, Arts and Humanities (miscellaneous), [INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG], Audio and Speech Processing (eess.AS), [INFO.INFO-SD]Computer Science [cs]/Sound [cs.SD], FOS: Electrical engineering, electronic engineering, information engineering, Electrical Engineering and Systems Science - Audio and Speech Processing
Abstract

This article is a survey on deep learning methods for single and multiple sound source localization. We are particularly interested in sound source localization in indoor/domestic environment, where reverberation and diffuse noise are present. We provide an exhaustive topography of the neural-based localization literature in this context, organized according to several aspects: the neural network architecture, the type of input features, the output strategy (classification or regression), the types of data used for model training and evaluation, and the model training strategy. This way, an interested reader can easily comprehend the vast panorama of the deep learning-based sound source localization methods. Tables summarizing the literature survey are provided at the end of the paper for a quick search of methods with a given set of target characteristics., Accepted for publication in The Journal of the Acoustical Society of America

Published
2022
Full Text
View/download PDF

148. Perception de parole chez l’enfant porteur d’implant(s) cochléaire(s) : Étude sur l’Auditory Verbal Therapy et la Langue française Parlée Complétée

Author

Van Bogaert, Lucie, Machart, Laura, Vilain, Anne, Helene, Loevenbruck, Laboratoire de Psychologie et NeuroCognition (LPNC ), Université Savoie Mont Blanc (USMB [Université de Savoie] [Université de Chambéry])-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA), GIPSA - Perception, Contrôle, Multimodalité et Dynamiques de la parole (GIPSA-PCMD), GIPSA Pôle Parole et Cognition (GIPSA-PPC), Grenoble Images Parole Signal Automatique (GIPSA-lab), Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP ), Université Grenoble Alpes (UGA)-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP ), Université Grenoble Alpes (UGA)-Grenoble Images Parole Signal Automatique (GIPSA-lab), Université Grenoble Alpes (UGA), and European Project: 860755,Marie Skłodowska-Curie Innovative Training Networks,Comm4CHILD(2020)

Subjects
Enfants, Perception de parole, Déficience auditive, Langue française Parlée Complétée (LfPC), French Cued Speech, [SCCO.NEUR]Cognitive science/Neuroscience, Speech perception, Cochlear implants, Auditory Verbal Therapy (AVT), [SCCO.LING]Cognitive science/Linguistics, Implants cochléaires, Children, Hearing impairment
Abstract

International audience; Hearing loss in children can lead to a delay in oral language development. Auditory Verbal Therapy (AVT) is a uni-sensory method that relies mainly on the auditory channel to enable deaf children with implants to develop good hearing skills. The French Cued Speech (CF) is a multimodal tool that disambiguate lip reading by adding a manual gesture. This study investigates the contribution of these two tools on speech perception in children with cochlear implants, using a lexicality judgment task. The results suggest similar performance in children with a good level of CF, children enrolled in an AVT program, and children with normal hearing, but weaker performance in children with a low level of CF. Thus, early and child-specific care is crucial for optimal phonological development.; La perte auditive chez l'enfant peut entrainer un retard sur le développement du langage oral. L'Auditory Verbal Therapy (AVT) est une méthode uni-sensorielle qui se base principalement sur le canal auditif pour permettre aux enfants sourds implantés de développer de bonnes capacités auditives. La Langue française Parlée Complétée (LfPC) est un outil multimodal qui permet de désambiguïser la lecture labiale par l'ajout d'un geste manuel. Cette étude s'intéresse à l'apport de ces deux outils sur la perception de parole chez l'enfant porteur d'implant cochléaire, à partir d'une tâche de jugement de lexicalité. Les résultats suggèrent des performances similaires chez les enfants avec un bon niveau de LfPC, les enfants inscrits dans un programme AVT, et les enfants normoentendants, mais des performances plus faibles chez les enfants avec un faible niveau de LfPC. Ainsi, une prise en charge précoce et adaptée à l'enfant est nécessaire pour le développement phonologique.

Published
2022
Full Text
View/download PDF

149. Les auto-encodeurs variationnels dynamiques et leur application à la modélisation de spectrogrammes de parole

Author

Girin, Laurent, Bie, Xiaoyu, Leglaive, Simon, Hueber, Thomas, Alameda-Pineda, Xavier, GIPSA - Cognitive Robotics, Interactive Systems, & Speech Processing (GIPSA-CRISSP), GIPSA Pôle Parole et Cognition (GIPSA-PPC), Grenoble Images Parole Signal Automatique (GIPSA-lab), Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP ), Université Grenoble Alpes (UGA)-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP ), Université Grenoble Alpes (UGA)-Grenoble Images Parole Signal Automatique (GIPSA-lab), Université Grenoble Alpes (UGA), Vers des robots à l’intelligence sociale au travers de l’apprentissage, de la perception et de la commande (ROBOTLEARN), Inria Grenoble - Rhône-Alpes, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria)-Université Grenoble Alpes (UGA), Institut d'Électronique et des Technologies du numéRique (IETR), Université de Rennes (UR)-Institut National des Sciences Appliquées - Rennes (INSA Rennes), Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-Nantes Université - pôle Sciences et technologie, Nantes Université (Nantes Univ)-Nantes Université (Nantes Univ), Université de Nantes, and ANR-19-P3IA-0003,MIAI,MIAI @ Grenoble Alpes(2019)

Subjects
speech signals modeling, [SPI]Engineering Sciences [physics], speech spectrograms, spectrogrammes de parole, modélisation des signaux de parole, auto-encodeurs variationnels dynamiques, dynamical variational autoencoders, analyse-resynthèse de la parole, speech analysis-resynthesis
Abstract

International audience; The Variational Autoencoder (VAE) is a powerful deep generative model that is now extensively used to represent high-dimensional complex data via a low-dimensional latent space learned in an unsupervised manner. In the original VAE model, input data vectors are processed independently. In recent years, a series of papers have presented different extensions of the VAE to process sequential data, that not only model the latent space, but also model the temporal dependencies within a sequence of data vectors and corresponding latent vectors, relying on recurrent neural networks. We recently performed a comprehensive review of those models and unified them into a general class called Dynamical Variational Autoencoders (DVAEs). In the present paper, we present this class of models and illustrate their high potential for modeling (spectrograms of) speech signals with speech analysis-resynthesis experiments.; L'auto-encodeur variationnel (AEV) est un modèle génératif profond permettant d'apprendre de façon auto-supervisé des représentations latentes compactes, à partir de données complexes de grande dimension. Dans le modèle AEV original, les vecteurs de données d'entrée sont traités indépendamment. Ces dernières années, plusieurs travaux ont proposé différentes extensions de l'AEV afin de traiter des données séquentielles (notamment temporelles). Ces modèles utilisent classiquement des réseaux de neurones récurrents pour tenir compte non seulement des dépendances entre les vecteurs d'une séquence d'entrée, mais également celles entre les représentations latentes correspondantes. Nous avons récemment effectué une revue complète de ces modèles et les avons unifiés en une classe générale appelée auto-encodeurs variationnels dynamiques (AEVDs). Dans le présent article, nous présentons cette classe de modèles et illustrons leur fort potentiel pour la modélisation des (spectrogrammes de) signaux de parole avec des expériences en analyse-resynthèse.

Published
2022
Full Text
View/download PDF

150. COSMO : un modèle bayésien des fondements sensorimoteurs de la perception et de la production de la parole

Author

Jean-Luc Schwartz, Pierre Bessière, Pascal Perrier, Marc-Antoine Georges, Mamady Nabé, Julien Diard, Marie-Lou Barnaud, Raphaël Laurent, Jean-François Patri, Clément Moulin-Frier, GIPSA - Perception, Contrôle, Multimodalité et Dynamiques de la parole (GIPSA-PCMD), GIPSA Pôle Parole et Cognition (GIPSA-PPC), Grenoble Images Parole Signal Automatique (GIPSA-lab), Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP ), Université Grenoble Alpes (UGA)-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP ), Université Grenoble Alpes (UGA)-Grenoble Images Parole Signal Automatique (GIPSA-lab), Université Grenoble Alpes (UGA), Institut des Systèmes Intelligents et de Robotique (ISIR), Sorbonne Université (SU)-Centre National de la Recherche Scientifique (CNRS), Laboratoire de Psychologie et NeuroCognition (LPNC ), Université Savoie Mont Blanc (USMB [Université de Savoie] [Université de Chambéry])-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA), Flowing Epigenetic Robots and Systems (Flowers), Inria Bordeaux - Sud-Ouest, Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria), Ce travail a bénéficié du soutien du Multidisciplinary Institute in Artificial Intelligence MIAI@Grenoble-Alpes (ANR-19-P3IA-0003) et du projet ISP Université Grenoble Alpes Bio-Bayes Predictions (ANR-15-IDEX-02, Auvergne-Rhône-Alpes (AURA) Region PAI-19-008112-01 grant)., and ANR-19-P3IA-0003,MIAI,MIAI @ Grenoble Alpes(2019)

Subjects
motor theories, perceptual-motor theories, perceptuo-motrices, [SCCO.COMP]Cognitive science/Computer science, Perception, théories auditives, production, auditory theories, motrices
Abstract

International audience; COSMO ("Communicating Objects by Sensory-Motor Operations") is a framework for jointly modeling speech perception and production by considering sensory-motor relations as a core component of its program. COSMO allows to jointly formalize two major theoretical frameworks in speech research that are auditory and motor theories – but also to integrate them within perceptual-motor theories. This leads to new perception models associating auditory processing and motor knowledge, and new speech motor control models oriented toward the achievement of multimodal sensory goals. We present the main results obtained with COSMO, and perspectives about temporal processing and deep learning implementation allowing to get closer to learning on real data.; Nous avons développé un cadre de modélisation des processus de la communication parlée, COSMO (« Communicating Objects by Sensory-Motor Operations »), qui s'applique à la fois aux modèles de perception et de production de parole en installant les relations sensori-motrices au coeur de son programme. COSMO permet de formaliser conjointement deux cadres théoriques majeurs des recherches sur la communication parlée, les théories auditives et motrices-mais aussi de les intégrer au sein de théories perceptuo-motrices. Ceci conduit ainsi à de nouveaux modèles de perception alliant traitements auditifs et prise en compte de connaissances motrices, ou de nouveaux modèles de contrôle moteur de la parole orientés vers la réalisation de buts sensoriels multimodaux. Nous présentons ces avancées ainsi que des pistes de développement sur le traitement temporel et l'implémentation deep learning permettant d'aller vers l'apprentissage sur des données réelles.

Published
2022

Catalog

Books, media, physical & digital resources
See catalog results