Research Interests:
International audienc
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests: Computer Science and NIST
Research Interests:
The IRIM group is a consortium of French teams working on Multimedia Indexing and Retrieval. This paper describes its participation to the TRECVID 2012 semantic indexing and instance search tasks. For the semantic indexing task, our... more
The IRIM group is a consortium of French teams working on Multimedia Indexing and Retrieval. This paper describes its participation to the TRECVID 2012 semantic indexing and instance search tasks. For the semantic indexing task, our approach uses a six-stages processing pipelines for computing scores for the likelihood of a video shot to contain a target concept. These scores are then used for producing a ranked list of images or shots that are the most likely to contain the target concept. The pipeline is composed of the following steps: descriptor extraction, descriptor optimization, classification, fusion of descriptor variants, higher-level fusion, and re-ranking. We evaluated a number of different descriptors and tried dierent fusion strategies. The best IRIM run has a Mean Inferred Average Precision of 0.2378, which ranked us 4th out of 16 participants. For the instance search task, our approach uses two steps. First individual methods of participants are used to compute simil...
Research Interests:
info:eu-repo/semantics/publishe
Research Interests:
The objective of this paper is to outline the design specification, implementation and evaluation of a proposed accelerated encryption framework which deploys both homomorphic and symmetric-key encryptions to serve the privacy preserving... more
The objective of this paper is to outline the design specification, implementation and evaluation of a proposed accelerated encryption framework which deploys both homomorphic and symmetric-key encryptions to serve the privacy preserving processing; in particular, as a sub-system within the Privacy Preserving Speech Processing framework architecture as part of the PPSP-in-Cloud Platform. Following a preliminary study of GPU efficiency gains optimisations benchmarked for AES implementation we have addressed and resolved the Big Integer processing challenges in parallel implementation of bilinear pairing thus enabling the creation of partially homomorphic encryption schemes which facilitates applications such as speech processing in the encrypted domain on the cloud. This novel implementation has been validated in laboratory tests using a standard speech corpus and can be used for other application domains to support secure computation and privacy preserving big data storage/processin...
Research Interests:
This presentation deals with different research axis followed by the Televigilance activity in Telecom Sudparis : a) Multi-sensors Ambulatory Terminal on Patient: Automatic Fall detector and noise-robust vital signal extraction, Alarm... more
This presentation deals with different research axis followed by the Televigilance activity in Telecom Sudparis : a) Multi-sensors Ambulatory Terminal on Patient: Automatic Fall detector and noise-robust vital signal extraction, Alarm automatisation and context identification based on Data Fusion, Physiological and actimetric data (cardiac frequency, movements,...) Fall, posture or activity recognition, Contextual information (sounds, localisation, activity) ANASON with ESIGETEL (D. Istrate) b) Cardiac pathology detection (R. Andreao): ECG signal segmentation based on sub-beat Hidden-Markov Models, Ischemias and Arrythmias events classification, Voice-controlled man-machine interface for Elderly persons to the smart home environment with TelecomParisTech (G. Chollet, P. Milhorat) c) Activity monitoring in care-houses in collaboration with LEGRAND company (P.Dore, T. Guettari) It also enlarges the scope to other teams within the Institut Mines-Telecom group, such as the QoL (Quality ...
Research Interests:
Research Interests: Engineering and Database
Cet article s'inscrit dans le cadre de la detection de mots cles dans un flux de parole. Nous presentons le probleme de detection comme un probleme de classification ou chaque mot cle peut appartenir a deux classes differentes, a... more
Cet article s'inscrit dans le cadre de la detection de mots cles dans un flux de parole. Nous presentons le probleme de detection comme un probleme de classification ou chaque mot cle peut appartenir a deux classes differentes, a savoir ``correct'' et ``incorrect''. Cette classification est realisee tout d'abord, par l'utilisation des Reseaux de Neurones Artificiels (RNA) en particulier le Perceptron Multi-Couches (PMC). Ensuite, nous proposons l'utilisation des SVM comme technique de classification innovante et efficace et qui a fait ses preuves dans plusieurs domaines de recherche. Chaque mot cle reconnu est represente par un vecteur caracteristique qui constitue l'entree du classifieur. Pour determiner ce vecteur, nous proposons trois representations vectorielles basees sur l'emploi des probabilites d'observations acoustiques locales et de la duree de chaque etat
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Dans ce rapport, nous abordons le probleme de la detection de mots cles dans un flux de parole et le rejet des entrees incorrectes.Nous proposons deux techniques differentes pour ameliorer le rejet des mots hors-vocabulaire. La premiere... more
Dans ce rapport, nous abordons le probleme de la detection de mots cles dans un flux de parole et le rejet des entrees incorrectes.Nous proposons deux techniques differentes pour ameliorer le rejet des mots hors-vocabulaire. La premiere est un modele combine qui utilise deux modeles ``poubelle'' (un modele appris et un modele determine au cours de la phase de reconnaissance). La deuxieme est une methode hybride basee sur un modele "poubelle" appris et une mesure de confiance calculee pour chaque hypothese de reconnaissance dans une etape de post-traitement.Ces deux approches sont evaluees dans un contexte d'une application boursiere. Nous avons etudie ces deux techniques afin de trouver les valeurs des parametres qui nous permettent d'ameliorer le taux de reconnaissance.
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Since life expectancy has increased significantly over the past century, society is being forced to discover innovative ways to support active aging and elderly care. The e-VITA project, which receives funding from both the European Union... more
Since life expectancy has increased significantly over the past century, society is being forced to discover innovative ways to support active aging and elderly care. The e-VITA project, which receives funding from both the European Union and Japan, is built on a cutting edge method of virtual coaching that focuses on the key areas of active and healthy aging. The requirements for the virtual coach were ascertained through a process of participatory design in workshops, focus groups, and living laboratories in Germany, France, Italy, and Japan. Several use cases were then chosen for development utilising the open-source Rasa framework. The system uses common representations such as Knowledge Bases and Knowledge Graphs to enable the integration of context, subject expertise, and multimodal data, and is available in English, German, French, Italian, and Japanese.
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
This paper outlines the EMPATHIC Research & Innovation project, which aims to research, innovate, explore and validate new interaction paradigms and platforms for future generations of Personalized Virtual Coaches to assist elderly people... more
This paper outlines the EMPATHIC Research & Innovation project, which aims to research, innovate, explore and validate new interaction paradigms and platforms for future generations of Personalized Virtual Coaches to assist elderly people living independently at and around their home. Innovative multimodial face analytics, adaptive spoken dialogue systems, and natural language interfaces are part of what the project investigates and innovates, aiming to help dependent aging persons and their carers. It will use remote, non-intrusive technologies to extract physiological markers of emotional states and adapt respective coach responses. In doing so, it aims to develop causal models for emotionally believable coach-user interactions, which shall engage elders and thus keep off loneliness, sustain health, enhance quality of life, and simplify access to future telecare services. Through measurable end-user validations performed in Spain, Norway and France (and complementary user evaluati...
Research Interests:
The EMPATHIC Research & Innovation project will research, innovate, explore and validate new paradigms and platforms, laying the foundation for future generations of Personalised Virtual Coaches to assist elderly people living... more
The EMPATHIC Research & Innovation project will research, innovate, explore and validate new paradigms and platforms, laying the foundation for future generations of Personalised Virtual Coaches to assist elderly people living independently at and around their home. Innovative multimodal face analytics, adaptive spoken dialogue systems and natural language interfaces are part of what the project will research and innovate, in order to help dependent aging persons and their carers. The project will use remote non-intrusive technologies to extract physiological markers of emotional states in real-time for online adaptive responses of the coach, and advance holistic modelling of behavioural, computational, physical and social aspects of a personalised expressive virtual coach. It will develop causal models of coach-user interactional exchanges that engage elders in emotionally believable interactions keeping off loneliness, sustaining health status, enhancing quality of life and simpli...
Research Interests:
Many elderly and dependent people living at home suffer from a lack of social contact. With their strength and physical condition decreasing, they are also reluctant to walk outside. Even in sheltered accommodation or hospitals, they may... more
Many elderly and dependent people living at home suffer from a lack of social contact. With their strength and physical condition decreasing, they are also reluctant to walk outside. Even in sheltered accommodation or hospitals, they may have phases of loneliness when they are left alone or when the personnel cannot continuously take care of them. With an aging population and the financial difficulties of having a full time caregiver for every dependent person living at home, the proliferation of advanced assistant robots seems to be a viable future solution. However, as most of what can be done with a robot is also possible without it, it is sometimes difficult to quantify the real value this technology can add to the current situation. Hence we believe that such a robot should be a reliable assistant, capable of helping a person indoors as well as outdoors. Furthermore, it should be a companion for dialoging, as well as a system capable of detecting health problems. The Roberta Ir...
Research Interests:
This paper presents an overview of a strategy for enabling speech recognition to be performed in the cloud whilst preserving the privacy of users. The strategy advocates a demarcation of responsibilities between the client and server-side... more
This paper presents an overview of a strategy for enabling speech recognition to be performed in the cloud whilst preserving the privacy of users. The strategy advocates a demarcation of responsibilities between the client and server-side components for performing the speech recognition task. On the client-side resides the acoustic model, which symbolically encodes the audio and encrypts the data before uploading to the server. The server-side then employs searchable encryption-based language modelling to perform the speech recognition task. The paper details the proposed client-side acoustic model components, and the proposed server-side searchable encryption which will be the basis of the language modelling. Some preliminary results are presented, and potential problems and their solutions regarding the encrypted communication between client and server are discussed. Preliminary benchmarking results with acceleration of the client and server operations with GPGPU computing are als...
Research Interests:
Research Interests:
Research Interests:
Cet article presente un systeme d’identification audio pour detecter et identifier des publicites et des morceaux de musique dans les flux radiophoniques en utilisant des unites acoustiques. Ces unites, nommees ALISP (Automatic Language... more
Cet article presente un systeme d’identification audio pour detecter et identifier des publicites et des morceaux de musique dans les flux radiophoniques en utilisant des unites acoustiques. Ces unites, nommees ALISP (Automatic Language Independent Speech Processing), sont apprises de maniere entierement automatique grâce a la decomposition temporelle, la quantification vectorielle et des modeles HMM. L’originalite de l’approche est qu’aucune transcription n’est utilisee pour apprendre les modeles HMM. Pour identifier des morceaux de musique et les publicites, les transcriptions ALISP des morceaux de reference sont comparees aux transcriptions du flux radiophonique de test en utilisant la distance de Levenshtein. Pour l’identification des publicites, nous obtenons un taux de precision de 99% et un taux de rappel de 94% pour un flux de test contenant 4401 publicites. Pour l’identification de morceaux de musique nous obtenons un taux de precision de 100% et un taux de rappel de 95% su...
Research Interests:
Nous decrivons une etude de la pertinence de HMMs parametriques, dans lesquels les parametres des lois gaussiennes dependent de variables externes dites contextuelles, pour une tâche de reconnaissance en environnement bruite. Les... more
Nous decrivons une etude de la pertinence de HMMs parametriques, dans lesquels les parametres des lois gaussiennes dependent de variables externes dites contextuelles, pour une tâche de reconnaissance en environnement bruite. Les resultats montrent d'une part l'interet de ce type de modelisation pour des variables contextuelles de differentes natures, des variables calculees a partir du signal lui-meme ou bien correspondant a des informations additionnelles sur le signal
Research Interests:
This paper presents the Intelligent Voice (IV) system submitted to the NIST 2016 Speaker Recognition Evaluation (SRE). The primary emphasis of SRE this year was on developing speaker recognition technology which is robust for novel... more
This paper presents the Intelligent Voice (IV) system submitted to the NIST 2016 Speaker Recognition Evaluation (SRE). The primary emphasis of SRE this year was on developing speaker recognition technology which is robust for novel languages that are much more heterogeneous than those used in the current state-of-the-art, using significantly less training data, that does not contain meta-data from those languages. The system is based on the state-of-the-art i-vector/PLDA which is developed on the fixed training condition, and the results are reported on the protocol defined on the development set of the challenge.
Research Interests:
Resume. vAssist est un projet du programme ‘Ambient Assisted Living’ de la communaute europeenne. Il propose le developpement et l’experimentation d’un assistant personnel centralise, le Majordome, pour les personnes dependantes. Ces... more
Resume. vAssist est un projet du programme ‘Ambient Assisted Living’ de la communaute europeenne. Il propose le developpement et l’experimentation d’un assistant personnel centralise, le Majordome, pour les personnes dependantes. Ces personnes utilisent un smartphone pour se connecter a un serveur. Elles dialoguent avec leurs Majordomes en Voix et Video sur IP. Le Majordome detecte les situations de detresse et contacte les aidants et services medicaux si necessaire. Cet article donne quelques details sur l’architecture de telecommunication, sur le systeme de dialogue vocal et des resultats d’evaluation du systeme de reconnaissance automatique de la parole.
Research Interests:
The main objective of the EMPATHIC project has been the design and development of a virtual coach to engage the healthy-senior user and to enhance well-being through awareness of personal status. The EMPATHIC approach addresses this... more
The main objective of the EMPATHIC project has been the design and development of a virtual coach to engage the healthy-senior user and to enhance well-being through awareness of personal status. The EMPATHIC approach addresses this objective through multimodal interactions supported by the GROW coaching model. The paper summarizes the main components of the EMPATHIC Virtual Coach (EMPATHIC-VC) and introduces a demonstration of the coaching sessions in selected scenarios.