EYE TRACKING AS A LANGUAGE PROFICIENCY TEST

Patent №

US 10,997,367

Granted

2021-05-04

Filed 2018

Owner

MASSACHUSETTS INSTITUTE OF TECHNOLOGY

AI components

3

ml · nlp · speech

Assignment

Recorded

Dataset

AIPD

2023_r1 edition

Application

16130627

In an embodiment, Applicant's method can automatically determine proficiency in a given language by tracking a user's gaze during reading a sample text. The language proficiency test includes reading sentences in a language (e.g., a language to the user's native language). The user's, or learner's, gaze is recorded using an eye-tracking camera while they read the sample text. Applicant's method and corresponding system predicts the language proficiency of the learner based on their gaze patterns. Applicant's method and corresponding system can also predict performance on specific standardized language proficiency tests such as Michigan EPT (Michigan English Proficiency Test), TOEIC® (Test of English for International Communication®), and TOEFL® (Test of English as a Foreign Language®).

Machine learningNatural languageSpeechG06F 40/263A61B 5/163A61B 5/7267G06F 3/005G06F 3/013G06F 17/18G06F 40/30G06N 3/09+9 more

AI classification

Natural language1.00
Machine learning1.00
Speech1.00
AI hardware0.11
Planning0.06
Vision0.05
Evolutionary computation0.00
Knowledge representation0.00

Ownership

MASSACHUSETTS INSTITUTE OF TECHNOLOGY

assignment · 484210812

© 2026 NYSGPT2525 LLC