Loading...
conference paper
On Use of Task Independent Training Data in Tandem Feature Extraction
2004
Proceedings of the 2004 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP-04)
The problem we address in this paper is, whether the feature extraction module trained on large amounts of task independent data, can improve the performance of stochastic models? We show that when there is only a small amount of task specific training data available, tandem features trained on task independent data give considerable improvement over Perceptual Linear Prediction (PLP) cepstral features in Hidden Markov Model (HMM) based speech recognition systems.
Loading...
Name
rr-03-57.pdf
Access type
openaccess
Size
217.13 KB
Format
Adobe PDF
Checksum (MD5)
a23e9a9042663c230295d222eb2bf5e3