An Overview and Discussion of the Suitability of Existing Speech Datasets to Train Machine Learning Models for Collective Problem Solving
Gnaneswar Villuri, Alex Doboli
TL;DR
The paper analyzes whether existing speech datasets designed for Spoken Language Understanding (SLU) adequately support training methods for Collaborative Problem Solving (CPS). It introduces a CPS-centric metric framework that covers cognitive, social, and emotional activities, and applies these metrics to a broad set of SLU datasets to assess their coverage of CPS-relevant phenomena. The findings reveal wide variability across SLU datasets in modalities, annotation depth, and coverage of CPS processes, with notable gaps in CPS-specific multi-modal and longitudinal data. The authors argue for new datasets that capture richer CPS dynamics—multi-modal signals, long-term team evolution, and ill-defined tasks—to better drive robust CPS-aware learning and analysis tools.
Abstract
This report characterized the suitability of existing datasets for devising new Machine Learning models, decision making methods, and analysis algorithms to improve Collaborative Problem Solving and then enumerated requirements for future datasets to be devised. Problem solving was assumed to be performed in teams of about three, four members, which talked to each other. A dataset consists of the speech recordings of such teams. The characterization methodology was based on metrics that capture cognitive, social, and emotional activities and situations. The report presented the analysis of a large group of datasets developed for Spoken Language Understanding, a research area with some similarity to Collaborative Problem Solving.