A novel Arabic Speech Recognition Method using Neural Networks and Gaussian Filtering

Authors: Nidal M. Turab, Khalaf Khatatneh, Ashraf Odeh

Abstract:
Phoneme recognition is a very popular area of research, especially in the field of machine intelligence, as it adds a distinguish touch in technology world in many fields that provides many features, This paper discussed a phoneme recognition using neural network technique by using a strong algorithm in pre-processing stage to get good results. Moreover, it explained the stages of phoneme recognition, pre-processing procedures such as getting a signal, sampling, quantization, determine an energy then using neural network to get a good results and enhancements. Gaussian Low Pass filtering was used to get enhanced results of voice signal and lowering the noisy; and then use neural network in a training stage to train a system to recognize the speech signal, this paper has benefited from some of the previous results for other researchers and entered good ideas to enhance Arabic voice recognition.

Keywords:
Neural networks, voice recognition, phoneme, wavelet, Gaussian filtering.


Volume 19, Issue 1, March 2014, pp. 865-869                    Download PDF



References:


[1] A. Waibel T. Hanazawa G. Hinton, IC Shikan and K.lang, "Phoneme Recognition: Neural Networks vs Hidden Markov Models".
[2] F. Freitag, & E. Monte “Phoneme recognition by means of predictive neural networks” Biological and Artificial Computation: From Neuroscience to Technology Lecture Notes in Computer Science Volume 1240, 1997, pp 1136-1143
[3] - Mohammed Rokibul Alam Kotwal,& Foyzul Hassan, “Recurrent Neural Network Based Phoneme Recognition Incorporating Articulatory Dynamic Parameters Advances in Computing and Communications Communications in Computer and Information Science Volume 192, 2011, pp 349-356
[4] Chiraz Jlassi, , Najet Arous and Noureddine Ellouze “Phoneme Recognition by Means of a Growing Hierarchical Recurrent Self-Organizing Model Based on Locally Adapting Neighborhood Radii” Cognitive Computation September 2010, Volume 2, Issue 3, pp 142-150
[5] N.Uma Maheswari, ,A.P.Kabilan and R.Venkatesh “SPEAKER INDEPENDENT PHONEME RECOGNITION USING NEURAL NETWORKS” Journal of Theoretical and Applied Information Technology Volume 2 no.6 pages 230-235
[6] Ghassan S. Mosa and adbuallah Ali “Arabic Phoneme recognition using Neural Fuzzy Petri Net and LPC Feature Extraction ” Signal Processing: An International Journal (SPIJ) Volume (3) : Issue (5) pages 161-171
[7] Mahmoud.A.Osman, Nasser Al, “Speech compression using LPC and wavelet” 2010 2nd International Conference on Computer Engineering and Technology volume 7 pages 93-99
[8] C.J.Long and S.Datta “Wavelet Based Feature Extraction for Phoneme Recognition”, http://www.asel.udel.edu/icslp/cdrom/vol1/239/a239.pdf
[9] K. Elenius and G. Takács “Phoneme Recognition Using Multi-Layer Perceptrons” http://www.speech.kth.se/prod/publications/files/1168.pdf
[10] Bernd T. Meyer, Tim Jürgens,” phoneme recognition depending on speech-intrinsic variability” J. Acoust. Soc. Am., Vol. 128, No. 5, November 2010, Pages: 3126–3141.
[11] K SREENIVASA RAO “Role of neural network models for developing speech systems” Sadhan ¯ a¯ Vol. 36, Part 5, October 2011, pp. 783–836.c Indian Academy of Sciences.
[12] BT Meyer ”Human and automatic speech recognition in the presence of speech-intrinsic variations” http://d-nb.info/1007410892/34.
[13] Randall Matignon ”Neural Network Modeling Using Sas Enterprise Miner” 2005
[14] Anil Kumar Vuppala “Neural Network based Feature Transformation for Emotion Independent Speaker Identification” Springer International Journal of Speech Technology Report No: IIIT/TR/2012/-1, pp335-349.
[15] O. FAROOQ, S. DATTA “wavelet sub-band based temporal features for robust hindi phoneme recognition” Int. J. Wavelets Multiresolut Inf. Process. Volume 08, Issue 06, November 2010
[16] N.UmaMaheswari,A.P.Kabilan , R.Venkatesh, "SPEAKER INDEPENDENT PHONEME RECOGNITION USING NEURAL NETWORKS",India, 2005 - 2009 JATIT
[17] L.Mesbahi and A.Benyetto,”Continuous speech recognition by adaptive temporal radial basis function,” in IEEE international conference on systems,Manand Cybernetics,2004,pp.574-579.
[18] Bernd T. Meyer, Tim Jürgens, Thorsten Wesker, Thomas Brand, and Birger Kollmeier, "Human phoneme recognition depending on speech-intrinsic variability", Medizinische Physik, Carl-von-Ossietzky Universität Oldenburg, D- 26111Oldenburg, Germany , September 2010.
[19]
ASM Shihavuddin, Mir Mohammad Nazmul Arefin, Mir Nahidul Ambia, Shah Ahsanul Haque, Tanvir Ahammad, "Development of real time Face detection system using Haar like features and Adaboost algorithm"
IJCSNS International Journal of Computer Science and Network Security, Vol: 10, Issue: 1, Page: 171-178
[20] Mahmood Yousefi Azar, Farbod Razzazi,"A neural predictive coding feature extraction scheme in DCT domain for phoneme recognition", 15 September 2010.
[21] C.J.Long and S.Datta, "Wavelet Based Feature Extraction for Phoneme Recognition", Loughborough LE11 3TU, UK.
[22] Wesfried, E; Wickerhauser, M.V. “Adapted local trigonometric transforms and speech processing,” IEEE SP41,3597-3600 (1993).
[23] Rafael C.Gonzalez, Richard E. Wood, "Digital Image Processing", third edition.
[24] K SREENIVASA RAO, "Role of neural network models for developing speech systems", Indian Institute of Technology Kharagpur.
[25] YIMING HUANG, Janusz A. Starzyk, Dennis Irwin, " Phoneme Recognition Using Neural Nework And Sequence Learning Model ", March 2009
.