Abstract
Automatic speech recognition (ASR) systems became an important part of our lives and are used by millions of people. However, scientists still try to improve their accuracy using many different techniques. In this paper, we focus on the influence the training set size has on the performance of Hidden Markov Model (HMM) based digit recognition system in Macedonian. The experiments are conducted using dataset consisting of 3093 samples divided in several different-sized training sets and one test set. Additionally, the behavior of several classification techniques was evaluated for the same issue. The best result was 19.9% error rate for 1500 samples in the training set using HMM based ASR system. This indicates that for this particular problem using the specified dataset the ideal number of samples for the training set is around 1500.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Juang, B.H., Rabiner, L.R.: Automatic Speech Recognition - A Brief History of the Technology Development. Rutgers University and the University of California, Santa Barbara (2004)
Plannerer, B.: An Introduction to Speech Recognition. ver. 1.1, Munich, Germany (2005)
Lippmann, R.P.: Neural Networks Classifiers for Speech Recognition. The Lincoln Laboratory Journal 1(1) (1988)
Rabiner, L.R.: Applications of Speech Recognitio in the Area of Telecommunications. AT&T Labs (1997)
Nagórski, A., Boves, L., Steeneken, H.: Optimal Selection of Speech Data For Automatic Speech Recognition Systems. Department of Language and Speech. University of Nijmegen, The Netherlands (2002)
Krajlevski, I., Mihajlov, D., Djordjevikj, D.: Hybrid Hmm/Ann System for Speech Recognition of Macedonian Language. In: Fifth National Conference With International Participation ETAI, Ohrid (2000)
Walker, W., Lamere, P., Kwok, P., Raj, B., Singh, R., Gouvea, E., Wolf, P., Woelfel, J.: Sphinx-4: A Flexible Open Source Framework for Speech Recognition. Sun Microsystems Inc. (2004)
Rabiner, L.R.: A Tutorial on Hidden Markov Models and Selected Applications in Seech Recognition. Proceedings of the IEEE 77(2) (1989)
University of Waikato, New Zealand: Machine learnings of softtware written in Java (Version 3.6) “Weka” (1997)
Madjarov, G.M.: Advanced methods for building hierarchical multi-label classifiers. Phd thesis, Faculty of Computer Science and Engineering, Skopje, Macedonia, pp. 48–50 (2012)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2015 Springer International Publishing Switzerland
About this paper
Cite this paper
Spasovski, D., Peshanski, G., Madjarov, G. (2015). The Influence the Training Set Size Has on the Performance of a Digit Speech Recognition System in Macedonian. In: Bogdanova, A., Gjorgjevikj, D. (eds) ICT Innovations 2014. ICT Innovations 2014. Advances in Intelligent Systems and Computing, vol 311. Springer, Cham. https://doi.org/10.1007/978-3-319-09879-1_21
Download citation
DOI: https://doi.org/10.1007/978-3-319-09879-1_21
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-09878-4
Online ISBN: 978-3-319-09879-1
eBook Packages: EngineeringEngineering (R0)