CloudySpeech

Speech production studies are presently served by several data acquisition methods that provide large amounts of data (e.g. electromagnetic midsagittal articulography, real-time magnetic resonance, ultrasound and surface electromyography). The main issue is that the field lacks the tools that would enable systematic and quantitative (possibly collaborative) exploration of such data. This project focuses on the research, development and usage of methods to support the storage, management, retrieval, processing and analysis of large sets of multilingual multimodal data for speech studies and on the deployment of such tools into Azure, allowing their on-demand use by the research community.

Keywords: multimodal speech production data, integrated data management, analysis as a service