Unsupervised Lexicon Discovery from Acoustic Input
Author(s)Lee, Chia-ying; O'Donnell, Timothy John; Glass, James R.
MetadataShow full item record
We present a model of unsupervised phonological lexicon discovery -- the problem of simultaneously learning phoneme-like and word-like units from acoustic input. Our model builds on earlier models of unsupervised phone-like unit discovery from acoustic data (Lee and Glass, 2012), and unsupervised symbolic lexicon discovery using the Adaptor Grammar framework (Johnson et al., 2006), integrating these earlier approaches using a probabilistic model of phonological variation. We show that the model is competitive with state-of-the-art spoken term discovery systems, and present analyses exploring the model's behavior and the kinds of linguistic structures it learns.
DepartmentMassachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory; Massachusetts Institute of Technology. Department of Brain and Cognitive Sciences
Transactions of the Association for Computational Linguistics
Association for Computational Linguistics
Lee, Chia-ying, Timothy J. O'Donnell, and James Glass. "Unsupervised Lexicon Discovery from Acoustic Input." Transactions of the Association for Computational Linguistics, Volume 3 (2015). © 2015 Association for Computational Linguistics
Final published version