Measuring the performance of isolated spoken Malay speech recognition using Multi-layer Neural Networks

• Main Authors: Bakar, N.A (Author), Bakar, Z.A (Author), Seman, N. (Author)
• Format: Article
• Language: English
• Subjects: Algorithms; Backpropagation; Back-propagation; Classification rates; Conjugate gradient algorithms; Conjugate gradient method; Data size; Hidden Neuron; Hidden neurons; Isolated word recognition; Levenberg-Marquardt algorithm; Malaysia; Mel-frequency cepstral coefficients; Melfrequency Cepstral Coefficients; Multi layer perceptron; Multi-layer Perceptron; Natural language processing systems; Network complexity; Network layers; Network structures; Neural networks; Recognition process; Speech extraction; Speech recognition; Speech recognition systems; Speech segmentation; Speech signal modeling; Spoken words; Stochastic errors; Training data; Two-layer network; Zero crossing rate
• Online Access: View Fulltext in Publisher; View in Scopus

 This paper describes speech signal modeling techniques which are suited to high performance and robust isolated word recognition. In this study, a speech recognition system is presented, specifically an isolated spoken Malay word recognizer which uses spontaneous and formally speeches collected from Parliament of Malaysia. Currently the vocabulary is limited to 25 words that can be pronounced exactly as it written and controls the distribution of the vocalic segments. The speech segmentation task is achieved by adopted energy based parameter and zero crossing rate measure with modification to better locates the beginning and ending points of speech from the spoken words. The training and recognition processes are realized by using Multi-layer Perceptron (MLP) Neural Networks with two-layer network configurations that are trained with stochastic error back-propagation to adjust its weights and biases after presentation of every training data. The Mel-frequency Cepstral Coefficients (MFCCs) has been chosen as speech extraction approach from each segmented utterance as characteristic features for the word recognizer. Recognition results showed that the performance of the two-layer networks increased as the numbers of hidden neurons increased. The best network structures average classification rate is 84.731% with (150-25) configuration. Implementation results also showed that the conjugate gradient (CG) algorithm was more accurate and reliable than the Levenberg-Marquardt (LM) algorithm for the network complexities and data size considered in this study. © 2010 IEEE.