Disordered Speech Data Collection: Lessons Learned at 1 Million Utterances from Project Euphonia
|Robert L. MacDonald (Google, USA), Pan-Pan Jiang (Google, USA), Julie Cattiau (Google, USA), Rus Heywood (Google, USA), Richard Cave (MND Association, UK), Katie Seaver (MGH Institute of Health Professions, USA), Marilyn A. Ladewig (Cerebral Palsy Associations of New York State, USA), Jimmy Tobin (Google, USA), Michael P. Brenner (Google, USA), Philip C. Nelson (Google, USA), Jordan R. Green (MGH Institute of Health Professions, USA), Katrin Tomanek (Google, USA)|
Speech samples from over 1000 individuals with impaired speech have been submitted for Project Euphonia, aimed at improving automated speech recognition systems for disordered speech. We provide an overview of the corpus, which recently passed 1 million utterances (>1300 hours), and review key lessons learned from this project. The reasoning behind decisions such as phrase set composition, prompted vs extemporaneous speech, metadata and data quality efforts are explained based on findings from both technical and user-facing research.