BabyCloud, a Technological Platform for Parents and Researchers

BabyCloud, a Technological Platform for Parents and Researchers Xuˆan-Nga Cao, Cyrille Dakhlia, Patricia Del Carmen, Mohamed-Amine Jaouani, Malik Ould-Arbi, Emmanuel Dupoux Laboratoire de Sciences Cognitives et Psycholinguistique, EHESS / Ecole Normale Sup´erieure / PSL Research University / CNRS / INRIA Paris, France ngafrance@gmail.com, dakhliacyrille@gmail.com, pat delcarmen@hotmail.com, jaouani.mohamed.amine@gmail.com, m.ouldarbi@gmail.com, emmanuel.dupoux@gmail.com Abstract In this paper, we present BabyCloud, a platform for capturing, storing and analyzing day-long audio recordings and photographs of children’s linguistic environments, for the purpose of studying infant’s cognitive and linguistic development and interactions with the environment. The proposed platform connects two communities of users: families and academics, with strong innovation potential for each type of users. For families, the platform offers a novel functionality: the ability for parents to follow the development of their child on a daily basis through language and cognitive metrics (growth curves in number of words, verbal complexity, social skills, etc). For academic research, the platform provides a novel means for studying language and cognitive development at an unprecedented scale and level of detail. They will submit algorithms to the secure server which will only output anonymized aggregate statistics. Ultimately, BabyCloud aims at creating an ecosystem of third parties (public and private research labs...) gravitating around developmental data, entirely controlled by the party whose data originate from, i.e. families. Keywords: language acquisition, cognitive development, linguistic development, day-long recordings, child speech corpus, child activity corpus 1. Introduction During their first years, infants develop both physically and cognitively at an amazing speed: many parents are eager to know more about this general process and about how well their child is doing. Recent progress in infant speech database collection (Xu et al., 2008; Roy, 2009; VanDam et al., 2016; Casillas et al., 2017; Warlaumont et al., 2017) and computational modeling of developmental processes (Mar- tin et al., 2016; Ludusan et al., 2015; Carbajal et al., 2016a; Carbajal et al., 2016b; Ludusan et al., 2017) open up possi- bilities to measure and model the progress of children using data collected in their natural environment. In other words, it becomes technically possible to offer parents scientifi- cally grounded analytics and tools to explore and document their child’s progress. In this paper, we introduce BabyCloud, a platform for cap- turing, storing and analyzing day-long audio and photo recordings of children’s linguistic environments. The plat- form is structured around 4 components: (a) Baby Log- ger which collects high definition pictures and high-quality recordings to be later decoded by automatic speech recog- nizers, (b) Baby Dock which transfers the data to a secure cloud server, recharges the recorder’s batteries, (c) Baby SmartBox, a secure cloud database which stores and in- dexes raw data and extracts metadata and developmental analytics (child’s linguistic landmarks, linguistic and social interactions clips and analyses) using machine learning al- gorithms and (d) Baby Explorer, which consists itself in two sub-components: a mobile application which allows families to manage and search their data, and an API acces- sible to scientists, will let them query summary statistics of the data and …

Content truncated to 500 words