Learning Label Embeddings for Nearest-Neighbor Multi-class Classification with an Application to Speech Recognition
Name
Collins_Learning label.pdf
Size
98.36 KB
Format
Adobe PDF
Checksum (MD5)
e2031d370cebb63995b820ebd611372d
Author(s)
Singh-Miller, Natasha
Collins, Michael
Date Issued
December 2009
Journal
23rd Annual Conference on Neural Information Processing Systems, 2009
Publisher
Neural Information Processing Systems (NIPS) Foundation
Citation
Singh-Miller, Natasha and Michael Collins. "Learning Label Embeddings for Nearest-Neighbor Multi-class Classification with an Application to Speech Recognition."
Version
Author's final manuscript
Abstract
We consider the problem of using nearest neighbor methods to provide a conditional
probability estimate, P(y|a), when the number of labels y is large and the
labels share some underlying structure. We propose a method for learning label
embeddings (similar to error-correcting output codes (ECOCs)) to model the similarity
between labels within a nearest neighbor framework. The learned ECOCs
and nearest neighbor information are used to provide conditional probability estimates.
We apply these estimates to the problem of acoustic modeling for speech
recognition. We demonstrate significant improvements in terms of word error rate
(WER) on a lecture recognition task over a state-of-the-art baseline GMM model.
MIT Department
Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory
Terms of Use
Attribution-Noncommercial-Share Alike 3.0 Unported
Persistent DSpace Link
