Academic Commons

Articles

Investigations into Tandem Acoustic Modeling for the Aurora Task

Ellis, Daniel P. W.; Reyes-Gomez, Manuel

In tandem acoustic modeling, signal features are first processed by a discriminantly-trained neural network, then the outputs of this network are treated as the feature inputs to a conventional distribution-modeling Gaussian-mixture model (GMM) speech recognizer. This arrangement achieves relative error rate reductions of 30% or more on the Aurora task, as well as supporting feature stream combination at the posterior level, which can eliminate more than 50% of the errors compared to the HTK baseline. In this paper, we explore a number of variations on the tandem structure to improve our understanding of its effectiveness. We experiment with changing the subword units used in each model (neural net and GMM), varying the data subsets used to train each model, substituting the posterior calculations in the neural net with a second GMM, and a variety of feature condition such as deltas, normalization and PCA rank reduction in the 'tandem domain' i.e. between the two models. All results are reported on the Aurora-2000 noisy digits task.

Files

Also Published In

Title
Eurospeech 2001: Scandinavia: 7th European Conference on Speech Communication and Technology: September 3-7, 2001, Aalborg Congress and Culture Centre, Aalborg-Denmark: proceedings

More About This Work

Academic Units
Electrical Engineering
Publisher
ISCA-Secretariat
Published Here
July 2, 2012
Academic Commons provides global access to research and scholarship produced at Columbia University, Barnard College, Teachers College, Union Theological Seminary and Jewish Theological Seminary. Academic Commons is managed by the Columbia University Libraries.