Modeling Infant Statistical Learning in Lexical Acquisition through Machine Learning
Jihee Hwang, Eric Ehizokhale and Krishan Kumar (CS221, Fall 2016)
The statistical learning model for infant language acquisition theorizes that children obtain lexical, grammatical and even phonological knowledge about a certain language by extracting statistical regularities from any input stream of words.
For this project, we replicated such lexical statistical learning given a set of speech audio inputs by incorporating techniques such as MFCC analysis, K-means clustering, and Uniform Cost Search.
Python was primarily used for the project.
Look below for the final report.
childlang_report