25-08-2017, 09:32 PM
Development of a Novel Voice Verification System using Wavelets
Novel Voice Verification.pdf (Size: 946.06 KB / Downloads: 76)
Abstract
This paper presents a novel voice verification
system using wavelet transforms. The conventional
signal processing techniques assume the signal to be
stationary and are ineffective in recognizing non
stationary signals such as the voice signals. Voice
signals which are more dynamic could be analyzed
with far better accuracy using wavelet transform.
The developed voice recognition system is word
dependant voice verification system combining the
RASTA and LPC. The voice signal is filtered using the
special purpose voice signal filter using the Relative
Spectral Algorithm (RASTA). The signals are de-
noised and decomposed to derive the wavelet
coefficients and thereby a statistical computation is
carried out. Further the formant or the resonance of
the voices signal is detected using the Linear
Predictive Coding (LPC). With the statistical
computation on the coefficients alone, the accuracy of
the verifying sample individual voice to his own voice
is quite high (around 75% to 80%). The reliability of
the signal verification is strengthened by combining
entailments from these two completely different aspects
of the individual voice. For voice comparison purposes
four out five individuals are verified and the results
show higher percentage of accuracy. The accuracy of
the system can be improved by incorporating advanced
pattern recognition techniques such as Hidden Markov
Model (HMM).
INTRODUCTION
The audio signal especially voice signal is
becoming one of the major part in human’s daily lives.
The basic function of the voice signal is that it is used
as one of the major tools for communication. However,
due to technological advancement, the voice signal is
further processed using software applications and the
voice signal information is utilized in various
application systems.
Discrete Wavelet Transform
Discrete Wavelet Transform (DWT) is a revised
version of Continuous Wavelet Transform(CWT). The
DWT compensates for the huge amount of data
generated by the CWT. The basic operation principles
of DWT are similar to the CWT however the scales
used by the wavelet and their positions are based upon
powers of two. This is called the dyadic scales and
positions as the term dyadic stands for the factor of two
[9]. As in many real world applications, most of the
important features of a signal lie in the low frequency
section. For voice signals, the low frequency content is
the section or the part of the signal that gives the signal
its identity whereas the high frequency content can be
considered as the part of the signal that gives nuance to
the signal. This is similar to imparting flavor to the
signal. For a voice signal, if the high frequency content
is removed, the voice will sound different but the
message can still be heard or conveyed. This is not true
if the low frequency content of the signal is removed as
what is being spoken cannot be heard except only for
some random noise.
SYSTEM IMPLEMENTATION
Variation System Implementation
In order to implement the system, a certain
methodology is implemented by decomposing the
voice signal to its approximation and detail. From the
approximation and detail coefficients that are
extracted, the methodology is implemented in order to
carry out the recognition process. The proposed
methodology for the recognition phase is the statistical
calculation. Four different types of statistical
calculations are carried out on the coefficients. The
statistical calculations that are carried out are mean,
standard deviation, variance and mean of absolute
deviation. The wavelet that is used for the system is the
symlet 7 wavelet as that this wavelet has a very close
correlation with the voice signal. This is determined
through numerous trial and errors. The coefficients that
are extracted from the wavelet decomposition process
is the second level coefficients as the level two
coefficients contain most of the correlated data of the
voice signal. The data at higher levels contains very
little amount of data deeming it unusable for the
recognition phase. Hence for initial system
implementation, the level two coefficients are used.
CONCLUSION
The Voice Recognition Using Wavelet Feature
Extraction employ wavelets in voice recognition for
studying the dynamic properties and characteristics of
the voice signal. This is carried out by estimating the
formant and detecting the pitch of the voice signal by
using LPC (Linear Predictive Coding). The voice
recognition system that is developed is word dependant
voice verification system used to verify the identity of
an individual based on their own voice signal using the
statistical computation, formant estimation and wavelet
energy. A GUI is built to enable the user to have an
easier approach in observing the step-by-step process
that takes place in Wavelet Transform. By using the
fifty preloaded voice signals from five individuals, the
verification tests have been carried and an accuracy
rate of approximately 80 % has been achieved. The
system can be enhanced further by using advanced
pattern recognition techniques such as Neural Network
or Hidden Markov Model (HMM) for more accurate
and efficient system.