RT Journal Article
T1 Data Balancing for Efficient Training of Hybrid ANN/HMM Automatic Speech Recognition Systems
A1 García-Moral, Ana I.
A1 Solera Ureña, R.
A1 Peláez Moreno, Carmen
A1 Díaz de María, Fernando
AB Hybrid speech recognizers, where the estimation of the emission pdf of the states of Hidden Markov Models (HMMs), usually carried out using Gaussian Mixture Models (GMMs), is substituted by Artificial Neural Networks (ANNs) have several advantages over the classical systems. However, to obtain performance improvements, the computational requirements are heavily increased because of the need to train the ANN. Departing from the observation of the remarkable skewness of speech data, this paper proposes sifting out the training set and balancing the amount of samples per class. With this method the training time has been reduced 18 times while obtaining performances similar to or even better than those with the whole database, especially in noisy environments. However, the application of these reduced sets is not straightforward. To avoid the mismatch between training and testing conditions created by the modification of the distribution of the training data, a proper scaling of the a posteriori probabilities obtained and a resizing of the context window need to be performed as demonstrated in the paper.
PB IEEE
SN 1558-7916
YR 2011
FD 2011-03
LK https://hdl.handle.net/10016/13074
UL https://hdl.handle.net/10016/13074
LA eng
NO This work was supported in part by the regional grant (Comunidad Autónoma de Madrid-UC3M) CCG06-UC3M/TIC-0812 and in part by a project funded by the Spanish Ministry of Science and Innovation (TEC 2008-06382).
DS e-Archivo
RD 1 may. 2024