8th International Conference on Spoken Language Processing

Jeju Island, Korea
October 4-8, 2004

Human Language Acquisition Methods in a Machine Learning Task

Nicole Beringer

IDSIA - Dalle Molle Institute for Artificial Intelligence, Switzerland

The goal of this study is to develop a psycho-computational model of human phoneme acquisition that includes the knowledge of linguistic universals to "teach" Artificial Neural Nets incrementally. Long Short-Term Memory (LSTM) artificial neural networks are capable to outperform previous recurrent networks on many tasks ranging from grammar recognition to speech and robot control. Together with our psycho-computational model they are supposed to recognize phonetic features in a way similar to humans learning to understand their first language.

Full Paper

Bibliographic reference.  Beringer, Nicole (2004): "Human language acquisition methods in a machine learning task", In INTERSPEECH-2004, 2233-2236.