Ahmed El-Shimi is the Co-Founder and Chief Product Officer of Speechace, a Seattle AI company dedicated to the advancement of spoken language assessment and feedback through Machine Learning. Ahmed holds a BSc. in Computer Engineering from Boston University and has worked in Machine Learning, Cloud and Infrastructure R&D for 20 years. Prior to Speechace, Ahmed spent 15 years at Microsoft in leadership roles at Product groups and Microsoft Research. His tenure spanned positions in Redmond (WA), Dublin (Ireland), Cairo (Egypt), and Mountain View (CA). Ahmed has spoken at dozens of academic and corporate conferences such as LinuxCon, SNIA Storage and Networking World, Huawei STW, and Microsoft Build. His work has been published at prestigious publications such as USENIX ATC and he holds multiple worldwide patents all of which in technologies that have been productized with commercial success.
17 November 2022 | 15:00 - 15:20 | English///Arabic | Using Machine Learning to evaluate Spoken Language Proficiency
We present a novel solution to speaking language assessment, one that is automated, can be scaled instantly, and can be applied effectively for placement and formative assessments in Education, or hiring assessments in the workplace. The AI-based spontaneous speaking assessment is conducted through a live authentic avatar that asks open-ended questions and evaluates speaking proficiency through state of the art speech and language assessment technology. We shall demonstrate how spoken language can be assessed in less than 10 minutes in the same manner a qualified human does it and discuss how Machine Learning is trained, validated, and used to comprehensively grade Fluency, Coherence, Pronunciation, Grammar, Vocabulary, and Relevance to establish the standard CEFR proficiency level of a candidate.