COMPARISON AND STATISTICAL ANALYSIS OF NAM AND NORMAL SPEECH PROCESSING USING WAVELET TRANSFORM
Journal Title: International Journal of Research in Computer and Communication Technology - Year 2013, Vol 2, Issue 4
Abstract
In this, we present statistical approaches to enhance body-conducted unvoiced speech for silent speech communication using wavelet transform. So far Analysis of NAM speech has been made only using HMM (Hidden Markov Model) and GMM (Gaussian Markov Model). In this paper, study of analyzing normal speech and NAM speech will be made by using Wavelet transform. Because, wavelets have the great advantage of being able to separate the fine details in a signal. It allows complex information to be decomposed into elementary forms at different positions and subsequently reconstruct it with high precision. Upon analyzing the Normal speech, NAM speech will also be analyzed using the same wavelet transform. The accuracy of words can be obtained from two methods that will be compared to determine the efficiency of using Wavelet transform in recognizing NAM speech. Wavelet analysis is capable of revealing aspects of data that other speech signal analysis technique such the extracted features are then passed to a classifier for the recognition of isolated words.
Authors and Affiliations
Janani P, Devi N
Keywords
Related Articles
- EP ID EP27560
- DOI -
- Views 319
- Downloads 5
Make Safe Data Promulgate In Multi Cloud Storage
The cloud computing technique provide a high range of service the society over the INTERNET. The unique features of clouding computing is that users data are processes remotely users does not own the machine or opera...
Robust Approach For Query Grouping Using User Search Logs
We proposes user’s are increasingly pursuing complex task orient goals on the Web such as making travel arrangements to managing finances or planning purchases. To better support users in their long term information...
Significance of Complementary Spectral Features for Speaker Recognition
Speaker recognition is the process of automatically recognizing who is speaking on the basis of individual information included in speech waves.. The most popular features for speaker recognition are Mel frequency ce...
Performance Analysis Of Vedic Multiplier Using Reversible Logic In Spartan 6
A system's performance is generally determined by the performance of the multiplier, because the multiplier is generally the slowest element in the system. Multipliers are key components of many high performance syst...
Defence Consideration of Strategy Classifiers under Assault
Pattern arrangement of action structures are for the most part used as a piece of hostile applications, as biometric affirmation, framework intrusion area, and spam isolating, in which data can be purposely controlle...