Sciweavers

Share
ICASSP
2010
IEEE

Multiple sequence alignment based bootstrapping for improved incremental word learning

9 years 12 months ago
Multiple sequence alignment based bootstrapping for improved incremental word learning
We investigate incremental word learning with few training examples in a Hidden Markov Model (HMM) framework suitable for an interactive learning scenario with little prior knowledge. When using only a few training examples the initialization of the models is a crucial step. In the bootstrapping approach proposed, an unsupervised initialization of the parameters is performed, followed by the retraining and construction of a new HMM using multiple sequence alignment (MSA). Finally we analyze discriminative training techniques to increase the separability of the classes using minimum classification error (MCE). Recognition results are reported on isolated digits taken from the TIDIGITS database.
Irene Ayllól Clemente, Martin Heckmann, Ger
Added 26 Jan 2011
Updated 26 Jan 2011
Type Journal
Year 2010
Where ICASSP
Authors Irene Ayllól Clemente, Martin Heckmann, Gerhard Sagerer, Frank Joublin
Comments (0)
books