This paper presents a study of automatic speech recognition system for Hindi utterances with regional Indian accents. In paper [3] we have designed matlab based ASR and control system for eight English key words by using simple rule base. This rule base algorithm is the beginning stage for Key Word recognition. In paper [4] we have designed Design of Hindi Key Word Recognition System for Home Automation System Using MFCC and DTW. Features of the speech signal are extracted in the form of MFCC coefficients and Dynamic Time Warping (DTW) has been used as features matching techniques. The recognition results are tested for clean and noisy test data. Average accuracy for clean data is 97.50 % while that for noisy data is 91.25 %. We face problem in noise environment to detect correct utterance now we are going to review different papers and find out different techniques to design our ASR control system for Hindi Key Words using MFCC and DTW in noise environment.
[1]
Louis A. Liporace,et al.
Maximum likelihood estimation for multivariate observations of Markov sources
,
1982,
IEEE Trans. Inf. Theory.
[2]
Abid M Jindani.
Speaker independent real-time speech recognition system
,
1998
.
[3]
Alex Acero,et al.
Spoken Language Processing: A Guide to Theory, Algorithm and System Development
,
2001
.
[4]
Abeer Alwan,et al.
HMM-Based Reconstruction of Unreliable Spectrographic Data for Noise Robust Speech Recognition
,
2010,
IEEE Transactions on Audio, Speech, and Language Processing.
[5]
J. Rajnoha,et al.
ASR systems in Noisy Environment : Analysis and Solutions for Increasing Noise Robustness
,
2011
.