I-Corps: Translation Potential of Speech-based Agents and Artificial Intelligence (AI) Assistive Tools in Language Learning

Award ID: 2524200

This project focuses on the development of speech-based artificial intelligence (AI) agents that are carefully designed using proven educational (pedagogical) principles. These AI agents act as assistive tools to help students improve their foreign language skills and enable extended, realistic conversational practice outside the classroom. Furthermore, this innovation can reduce instructors preparation time by up to 80% while increasing learners speaking practice time by as much as 90%. This technology has the potential to significantly strengthen foreign language acquisition and increase access to foreign language instruction, supporting skills that are vital to the strategic interests of the U.S. This I-Corps project utilizes experiential learning coupled with a first-hand investigation of the industry ecosystem to assess the translation potential of the technology. This technology addresses foreign language learning obstacles related to limited class time, student anxiety, and limited opportunities for practice, which hinder the development of foreign language skills necessary for basic conversation. Unlike existing tools, this technology utilizes recent innovations in generative artificial intelligence (GenAI) including large language models (LLMs) and large speech models (speech-to-text STT and text-to-speech TTS technologies) to design speech-based, language-agnostic agents that align with and augment classroom pedagogy. This technologys low cost and ease of integration represent a potentially important advance by providing cost-effective, scalable solutions for U.S. education institutions and students mastering new language skills. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Program Director: Ruth Shuman