Seminars home
Video-ethnographic studies of the interaction with and the use of digital voice assistants at home
at
NAVER LABS Europe is hosting a guest from Orange Lab who will speak about their respective work on visual localization.
The seminar run from 11am to 12pm. Please register online
Date: 24th May 2019
Speaker: Moustafa Zouinar, Researcher & associate Professor at Orange & CNAM, Paris, France.
Abstract: Since 2015, we are witnessing a massive development and diffusion of domestic Voice Assistants (VA) like Amazon Echo or Google Home. VA, which take the form of speakers that include a voice user interface, are presented by their promoters as systems that can facilitate domestic inhabitants’ daily lives through the use of speech. These devices offer a large number of services, both in-house and/or third-party. For example, they allow users to play media content (music, radio, video), search for information on internet, or receive news updates by interacting with the device through voice. These devices also have integrated support for home automation (control of home appliances like light, heating systems, etc.). This diffusion raises a number of questions about the usage of these devices: How do house occupants interact with them? How do they appropriate them? How do they use them and what for? How do VA affect their daily activities?
The presentation will shed some light on these questions, on the basis of a series of naturalistic video-ethnographic research studies that were conducted between 2015 and 2018 in France. A first study dealt with the study of the use of the first English-speaking assistants launched on the American market in 2015 (Amazon Echo and Ivee). A second study focused on a French-speaking experimental VA prototype developed by Orange. In these first studies, home participants were asked to test the devices during several weeks. The release in France of Google Home in 2017, then Alexa in 2018, allowed us in a third study to extend the research to households that had acquired these assistants (following a purchase or an offer). These studies enabled us to constitute a unique and large corpus of video data that show how people use and interact with VA in their homes. These data have been supplemented by interviews without and with self-confrontation to the videos. Overall, 23 homes participated to these studies. Drawing on conversational analysis, the presentation will examine the structure and the dynamics of the interactions between the users and the VAs. It will also show how the house occupants appropriate these devices through time. Another issue that will be examined has to do with the consequences of the use of VAs on the organization of the occupants’ practices at a collective level.
To make robots autonomous in real-world everyday spaces, they should be able to learn from their interactions within these spaces, how to best execute tasks specified by non-expert users in a safe and reliable way. To do so requires sequential decision-making skills that combine machine learning, adaptive planning and control in uncertain environments as well as solving hard combinatorial optimisation problems. Our research combines expertise in reinforcement learning, computer vision, robotic control, sim2real transfer, large multimodal foundation models and neural combinatorial optimisation to build AI-based architectures and algorithms to improve robot autonomy and robustness when completing everyday complex tasks in constantly changing environments.
VISION
The research we conduct on expressive visual representations is applicable to visual search, object detection, image classification and the automatic extraction of 3D human poses and shapes that can be used for human behavior understanding and prediction, human-robot interaction or even avatar animation. We also extract 3D information from images that can be used for intelligent robot navigation, augmented reality and the 3D reconstruction of objects, buildings or even entire cities.
Our work covers the spectrum from unsupervised to supervised approaches, and from very deep architectures to very compact ones. We’re excited about the promise of big data to bring big performance gains to our algorithms but also passionate about the challenge of working in data-scarce and low-power scenarios.
Furthermore, we believe that a modern computer vision system needs to be able to continuously adapt itself to its environment and to improve itself via lifelong learning. Our driving goal is to use our research to deliver embodied intelligence to our users in robotics, autonomous driving, via phone cameras and any other visual means to reach people wherever they may be.
This web site uses cookies for the site search, to display videos and for aggregate site analytics.
Learn more about these cookies in our privacy notice.
Cookie settings
You may choose which kind of cookies you allow when visiting this website. Click on "Save cookie settings" to apply your choice.
This content is currently blocked. To view the content please either 'Accept social media cookies' or 'Accept all cookies'.
For more information on cookies see our privacy notice.