Project Login
Registration No:
Password:
MAIL ALERTS SMS ALERTS
 
     
   
     

HISTOGRAM EQUALIZATION TO MODEL ADAPTATION FOR ROBUST SPEECH RECOGNITION

Platform : DSP

ABSTRACT We propose a new model adaptation method based on the histogram equalization technique for providing robustness in noisy environments. The trained acoustic mean models of a speech recognizer are adapted into environmentally matched conditions by using the histogram equalization algorithm on a single utterance basis. For more robust speech recognition in the heavily noisy conditions, trained acoustic covariance models are efficiently adapted by the signal-to-noise ratio-dependent linear interpolation between trained covariance models and utterance-level sample covariance models. Speech recognition experiments on both the digit-based Aurora2 task and the large vocabulary-based task showed that the proposed model adaptation approach provides significant performance improvements compared to the baseline speech recognizer trained on the clean speech data. PROJECT DESCRIPTION / ALGORITHM The application of HEQ to feature compensation begins with such an assumption [18] that the acoustic mismatch between clean reference (or training) features and noisy test features results in the statistical difference between their corresponding probability density functions (PDFs). Then, the idea of HEQ for feature compensation (HEQ-FC) is to conduct a transformation that converts the PDF of the original or test features into that of reference or training features to reduce the effects of noise corruption. In practice, reference and test PDFs are replaced by their corresponding histograms and the test histogram is equalized by using the transformation given by the HEQ algorithm [10]. Here, we assume that HEQ-FC is applied to each feature on a component-by-component basis for algorithmic simplicity. This assumption can be well accepted in the orthogonal transformation-based features such as cepstral features due to their low correlation. In this case, the algorithm of HEQ-FC is described as follows [10]. For given reference and test random features x and y, respectively, a transformation function of HEQ-FC mapping test PDF PY(y) into reference PDF PX(x) is obtained by equating their corresponding CDFs defined as (1) (2) where is the inverse of the reference CDF is the test CDF, and F(y) is the transformation function of HEQ-FC and has single valued, monotonically nondecreasing characteristics. OBJECTIVE: Histogram equalization algorithm on a single utterance to obtain robustness for speech recognition in noisy environments INPUT: utterances’ of speech signals OUTPUT : matched or speech uttrences APPLICATIONS: Authentications in different sectors SOFTWARE TOOL USED : MATLAB 2007A INTRODUCTION TO MATLAB MATLAB is a high-performance language for technical computing. It integrates computation, visualization, and programming in an easy-to-use environment where problems and solutions are expressed in familiar mathematical notation. Typical uses include: • Math and computation • Algorithm development • Modeling, simulation, and prototyping • Data analysis, exploration, and visualization • Scientific and engineering graphics • Application development, including Graphical User Interface building MATLAB is an interactive system whose basic data element is an array that does not require dimensioning. This allows you to solve many technical computing problems, especially those with matrix and vector formulations, in a fraction of the time it would take to write a program in a scalar noninteractive language such as C or Fortran.

NOW GET PROJECTS ! GET TRAINED ! GET PLACED !

IEEE, NON-IEEE, REAL TIME LIVE ACADEMIC PROJECTS,

PROJECTS WITH COMPLETE COURSES,SOFT SKILLS & PLACEMENTS

ALLOVER INDIA & WORLD WIDE,

HOSTEL FACILITY AVAILABLE FOR GIRLS & BOYS SEPARATELY,

CALL: 08985129129 ,  E-Mail Id: support@ascentit.in

REGISTER FOR PROJECTS NOW ! GET DISCOUNT
   
1