Comparative Study of Isolated Word Recognition System for Hindi Language

Speech is a natural way of communication and it provides an easy user interface to machines so that automatic speech recognition (ASR) system is considered as necessary. But the automatic speech recognition (ASR) system doesn’t perform perfectly for any language. So that generation of an accurate and robust acoustic model is necessary. The overall performance of any speech recognition system is highly depends on the feature extraction technique and classifier. In this paper, we have described Comparative study Isolated Word Recognition System for Hindi Language using MFCC as feature extraction and KNN as pattern classification technique. When our system is trained for First 10 words it achieves 89% rate of recognition and when trained for all 100 words it achieves 62.50% rate of recognition. As vocabulary increases performance decreases. Keywords— Feature extraction, MFCC, DCT, FFT, vocabulary, KNN .

[1]  Virender Kadyan,et al.  Punjabi Automatic Speech Recognition Using HTK , 2012 .

[2]  Parneet Kaur,et al.  Hindi Automatic Speech Recognition Using HTK , 2013 .

[3]  Roger F. Woods,et al.  Optimization of Weighted Finite State Transducer for Speech Recognition , 2013, IEEE Transactions on Computers.

[4]  Navdeep Singh,et al.  Literature Review on Automatic Speech Recognition , 2012 .

[5]  Ramesh R. Manza,et al.  Recognition of isolated words using Zernike and MFCC features for audio visual speech recognition , 2014, International Journal of Speech Technology.

[6]  Geeta Nijhawan,et al.  ISOLATED SPEECH RECOGNITIONUSING MFCC AND DTW , 2013 .

[7]  Ratnadeep R. Deshmukh,et al.  Indian Language Speech Database: A Review , 2012 .

[8]  Naveen Kumar,et al.  Automatic Speech Recognition System for Hindi Utterances with Regional Indian Accents: A Review , 2013 .

[9]  Mayank Dave,et al.  Integration of multiple acoustic and language models for improved Hindi speech recognition system , 2012, Int. J. Speech Technol..

[10]  A. V. Kulkarni,et al.  Overview: Speech Recognition Technology, Mel- frequency Cepstral Coefficients (MFCC), Artificial Neural Network (ANN) , 2013 .

[11]  Navpreet Singh,et al.  Speaker Accent Recognition by MFCC Using K- Nearest Neighbour Algorithm: A Different Approach , 2015 .

[12]  M. A. Anusuya,et al.  Speech Recognition by Machine, A Review , 2010, ArXiv.

[13]  Yogesh Kumar,et al.  Ensemble Feature Extraction Modules for Improved Hindi Speech Recognition System , 2012 .

[14]  Ratnadeep R. Deshmukh,et al.  A Review on Different Approaches for Speech Recognition System , 2015 .

[15]  Shailesh Khaparkar,et al.  Performance Improvement of Speaker Recognition System , 2012 .

[16]  Jafreezal Jaafar,et al.  FEATURE EXTRACTION USING MFCC , 2013 .

[17]  Dr. S V Sathyanarayana,et al.  Automatic Speech Recognition-A Literature Survey on Indian languages and Ground Work for Isolated Kannada Digit Recognition using MFCC and ANN , 2015 .

[18]  P Punitha,et al.  Speech Recognition Technology: A Survey on , 2013 .

[19]  D.R. Reddy,et al.  Speech recognition by machine: A review , 1976, Proceedings of the IEEE.

[20]  M. Kalamani,et al.  Automatic Speech Recognition using ELM and KNN Classifiers , 2015 .

[21]  Santosh Chapaneri,et al.  Spoken Digits Recognition using Weighted MFCC and Improved Features for Dynamic Time Warping , 2012 .

[22]  Tsang-Long Pao,et al.  A Weighted Discrete KNN Method for Mandarin Speech and Emotion Recognition , 2008 .

[23]  S. B. Magre.  A Review on Feature Extraction and Noise Reduction Technique , 2014 .

[24]  Ankit Kumar,et al.  Continuous Hindi Speech Recognition using Monophone based Acoustic Modeling , 2014 .