Speech Query Recognition in Tamil Language Using Wavelet and Wavelet Packets
0
조회
0
다운로드
3
인용
0
참고문헌
키워드
초록
Speech recognition is one of the fascinating fields in the area of Computer science. Accuracy of speechrecognition system may reduce due to the presence of noise present in speech signal. Therefore noise removalis an essential step in Automatic Speech Recognition (ASR) system and this paper proposes a new techniquecalled combined thresholding for noise removal. Feature extraction is process of converting acoustic signalinto most valuable set of parameters. This paper also concentrates on improving Mel Frequency CepstralCoefficients (MFCC) features by introducing Discrete Wavelet Packet Transform (DWPT) in the place ofDiscrete Fourier Transformation (DFT) block to provide an efficient signal analysis. The feature vector isvaried in size, for choosing the correct length of feature vector Self Organizing Map (SOM) is used. As asingle classifier does not provide enough accuracy, so this research proposes an Ensemble Support VectorMachine (ESVM) classifier where the fixed length feature vector from SOM is given as input, termed asESVM_SOM. The experimental results showed that the proposed methods provide better results than theexisting methods.
Abstract
Speech recognition is one of the fascinating fields in the area of Computer science. Accuracy of speechrecognition system may reduce due to the presence of noise present in speech signal. Therefore noise removalis an essential step in Automatic Speech Recognition (ASR) system and this paper proposes a new techniquecalled combined thresholding for noise removal. Feature extraction is process of converting acoustic signalinto most valuable set of parameters. This paper also concentrates on improving Mel Frequency CepstralCoefficients (MFCC) features by introducing Discrete Wavelet Packet Transform (DWPT) in the place ofDiscrete Fourier Transformation (DFT) block to provide an efficient signal analysis. The feature vector isvaried in size, for choosing the correct length of feature vector Self Organizing Map (SOM) is used. As asingle classifier does not provide enough accuracy, so this research proposes an Ensemble Support VectorMachine (ESVM) classifier where the fixed length feature vector from SOM is given as input, termed asESVM_SOM. The experimental results showed that the proposed methods provide better results than theexisting methods.
참고문헌 (25)
- A. G. Chitu, "Comparison between different feature extraction techniques for audio-visual speech recognition", Journal on Multimodal User Interfaces, 1(1), pp.7~20 (2007)
- R. Aggarwal, "Noise reduction of speech signal using wavelet transform with modified universal threshold", International Journal of Computer Applications, 20(5), pp.14~19 (2011)
- R. Sarikaya, "Wavelet packet transform features with application to speaker identification", Proceedings of 3rd IEEE Nordic Signal Processing Symposium, pp.81~84 (1998)
- J. N. Gowdy, "Mel-scaled discrete wavelet coefficients for speech recognition", Proceedings of the IEEE International Conference on Acoustics, Speech, and Signal Processing, pp.1351~1354 (2000)
- N. S. Nehe, "DWT and LPC based feature extraction methods for isolated word recognition", EURASIP Journal on Audio, Speech, and Music Processing, 2012(1), pp.1~7 (2012)
- B. Bharathi, "A neural network based speech recognition system for isolated Tamil words", Proceedings of International Conference on Neural Networks and Artificial Intelligence (2006)
- S. Saraswathi, "Morpheme based language model for Tamil speech recognition system", International Arab Journal of Information Technology, 4(3), pp.214~219 (2007)
- R. Arun Thilak, "Speech recognizer for Tamil language", Proceedings of Tamil Internet Conference, pp.1~7 (2004)
- V. Radha, "Isolated word recognition system for Tamil spoken language using back propagation neural network based on LPCC features", Computer Science & Engineering, 1(4), pp.1~11 (2011)
- I. Patel, "Speech recognition using HMM with MFCC : an analysis using frequency spectral decomposition technique", Signal & Image Processing : An International Journal(SIPIJ), 1(2), pp.101~110 (2010)
- "Tamil speech recognition: a complete model, Electronic Journal Technical Acoustics" (2008)[URL]
- S. Rojathai, "A novel speech recognition system for Tamil word recognition based on MFCC and FFBNN", European Journal of Scientific Research, 85(4), pp.578~590 (2012)
- A. N. Sigappi, "Spoken word recognition strategy for Tamil language", International Journal of Computer Science Issues, 9(1), pp.1694~0814 (2012)
- P. Sivaraj, "Recognition of isolated spoken words using DWT", International Journal of Engineering & Science Research, 2(9), pp.1187~1196 (2012)
- R. Thangarajan, "Word and triphone based approaches in continuous speech recognition for Tamil language", WSEAS Transactions on Signal Processing, 4(3), pp.76~86 (2008)
- S. Saraswathi, "Design of language models at various phases of Tamil speech recognition system", International Journal of Engineering, Science and Technology, 2(5), pp.244~257 (2010)
- S. Karpagavalli, "Isolated Tamil digits speech recognition using vector quantization", International Journal of Engineering Research and Technology, 1(4), pp.1~12 (2012)
- P. Iswarya, "Speech based query processing architecture for Tamil–English in cross language text retrieval system", International Journal of Emerging Trends in Engineering and Development, 7(2), pp.437~442 (2012)
- D. L. Donoho, "Minimax estimation via wavelet shrinkage", Annals of Statistics, 26(3), pp.879~921 (1998)
- P. Iswarya, "Comparative analysis of feature extraction techniques for Tamil speech recognition", Proceedings of International Conference on Emerging Research in Computing, Information, Communication and Application, pp.755~761 (2013)
- A. Ekbal, "Simulated annealing based classifier ensemble techniques : Application to part of speech tagging", Information Fusion, 14(3), pp.288~300 (2013)
- A. R. Ahmad, "Machine learning using support vector machines", Centre for Artificial Intelligence and Robotics, ~ (2002)
- "A user’s guide to support vector machines"[URL]
- H. C. Kim, "Constructing support vector machine ensemble", Pattern Recognition, 36(12), pp.2757~2767 (2003)
- "Speech recognition using neural networks"[URL]