Speaker identification and verification over short

distance telephone lines using artificial neural networksSPEAKER IDENTIFICATION AND VERIFICATION OVER SHORT
DISTANCE TELEPHONE LINES USING ARTIFICIAL NEURAL
NETWORKS
Ganesh K Venayagamoorthy, Narend Sunderpersadh, and Theophilus N Andrew
emailprotected emailprotected emailprotected
Electronic Engineering Department,
M L Sultan Technikon,
P O Box 1334, Durban, South Africa.

ABSTRACT
Crime and corruption have become rampant today
in our society and countless money is lost each year
due to white collar crime, fraud, and embezzlement.

This paper presents a technique of an ongoing work
to combat white-collar crime in telephone
transactions by identifying and verifying speakers
using Artificial Neural Networks (ANNs). Results
are presented to show the potential of this technique.

1. INTRODUCTION
Several countries today are facing rampant crime and
corruption. Countless money is lost each year due to
white collar crime, fraud, and embezzlement. In today’s
complex economic times, businesses and individuals
are both falling victims to these devastating crimes.

Employees embezzle funds or steal goods from their
employers, then disappear or hide behind legal issues.

Individuals can easily become helpless victims of
identity theft, stock schemes and other scams that rob
them of their money
White collar crime occurs in the gray area where the
criminal law ends and civil law begins. Victims of
white collar crimes are faced with navigating a daunting
legal maze in order to effect some sort of resolution or
recovery. Law enforcement is often too focused on
combating “street crime” or does not have the expertise
to investigate and prosecute sophisticated fraudulent
acts. Even if criminal prosecution is pursued, a criminal
conviction does not mean that the victims of fraud are
able to recover their losses. They have to rely on th
criminal courts awarding restitution after the conviction
and by then the perpetrator has disposed of or hidde
most of the assets available for recovery. From the civil
law perspective, resolution and recovery can just be a
difficult as pursuing criminal prosecution. Perpetrators
of white collar crime are often difficult to locate and
served with civil process. Once the perpetrators have
been located and served, proof must be provided that
the fraudulent act occurred and recovery/damages are
needed. This usually takes a lengthy legal fight, which
often can cost the victim more money than the fraud
itself. If a judgement is awarded, then the task of
collecting is made difficult by the span of time passed
and the perpetrator’s efforts to hide the assets. Often
after a long legal battle, the victims are left with a
worthless judgement and no recovery.

One solution to avoid white collar crimes and shorten
the lengthy time in locating and serving perpetrators
with a judgement is by the use of biometrics techniques
for identifying and verifying individuals. Biometrics are
methods for recognizing a user based on his/her unique
physiological and/or behavioural characteristics. These
characteristics include fingerprints, speech, face, retina,
iris, hand-written signature, hand geometry, wrist veins,
etc. Biometric systems are being commercially
developed for a number of financial and securit
applications.

Many people today have access to their company’s
information systems by logging in from home. Also,
internet services and telephone banking are widely used
by the corporate and private sectors. Therefore to
protect one’s resources or information with a simple
password is not reliable and secure in the world of
today. The conventional methods of using keys, access
passwords and access cards are being easily overcome
by people with criminal intention.

Voice signals as a unique behavioral characteristics is
proposed in this paper for speaker identification and
verification over short distance telephone lines using
artificial neural networks. This will address the white
collar crimes over the telephone lines. Speaker
identification 1 and verification 2 over telephone
lines have been reported but not using artificial neural
networks.

Artificial neural networks are intelligent systems that
are related in some way to a simplified biological model
of the human brain. Attenuation and distortion of voice
signals exist over the telephone lines and artificial
neural networks, despite a nonlinear, noisy and
unstationary environment, are still good at recognizing
and verifying unique characteristics of signals. Multilayer
perceptron (MLP) feedforward neural networks
trained with backpropagation algorithm have been
applied to identify bird species using recordings of
birdsongs 3. Speaker identification based on direct
voice signals using different types of neural networks
have been reported 4,5. The work reported in this
paper extends the work reported in 5 to short distance
telephone networks using ANN architectures described
in section 4 of this paper.

The feature extraction, the neural network architectures
and the software and hardware involved in the
development of the speaker identification and
verification system are described in this paper. Results
with success rates up to 90% in speaker identification
and verification over short distance telephone lines
using artificial neural networks is reported in this paper.

2. SPEAKER IDENTIFICATION AND
VERIFICATION SYSTEM
A block diagram of a conventional speaker
identification/verification system is shown in figure 1.

The system is trained to identify a person’s voice by
each person speaking out a specific utterance into the
microphone. The speech signal is digitized and some
digital signal processing is carried out to create a
template for the voice pattern and this is stored in
memory.

The system identifies a speaker by comparing the
utterance with the respective template stored in th
memory. When a match occurs the speaker is identified.

The two important operations in an identifier are the
parameter extraction and pattern matching. In paramete
extraction distinct patterns are obtained from the
utterances of each person and used to create a template.

In pattern matching, the templates created in the
parameter extraction process are compared with those
stored in memory. Usually correlation techniques are
employed for traditional pattern matching.

ADC Parameter
Extraction
Pattern
Matching
Memory
Template
Output
Device
mic
Figure 1: Block Diagram of a Conventional Speaker
Identification/Verification System.

The speaker identification/verification system over
telephone lines investigated in this paper using artificial
neural networks is shown in figure 2.

Feature
Extraction
Neural Network
Classification
Speaker Identity
or
Speaker Authenticity
Telephone
Speech Signal
Figure 2: Block Diagram of the Speaker
Identification/Verification System using an ANN.

In this paper, the speaker identification/verification
system reported is a text-dependent type. The system is
trained on a group of people to be identified by each
person speaking out the same phrase. The voice is
recorded on a standard 16-bit computer sound card from
the telephone handset receiver. Although the frequenc
of the human voice ranges from 0 kHz to 20 kHz, most
of the signal content lies in the 0.3 kHz to 4 kHz range.

The frequency over the telephone lines is limited to 0.3
kHz to 3.4 kHz and this is the frequency band of interest
in this work. Therefore, a sampling rate of 16 kHz
satisfying the Nyquist criterion is used. The voices are
stored as sound files on the computer. Digital signal
processing techniques are used to convert these sound
files to a presentable form as input vectors to a neural
network. The output of the neural network identifies
and verifies the speaker in the group.

3. FEATURE EXTRACTION
The process of feature extraction consists of obtaining
characteristic parameters of a signal to be used to
classify the signal. The extraction of salient features is a
key step in solving any pattern recognition problem. Fo
speaker recognition, the features extracted from a
speech signal should be consistent with regard to the
desired speaker while exhibiting large deviations from
the features of an imposter. The selection of speakerunique
features from a speech signal is an ongoing
issue. Findings report that certain features yield bette
performance for some applications than do other
features. Ref. 5 have shown on how the performance
can be improved by combining different types of
features as inputs to an ANN classifier.

Speaker identification and verification over telephone
network presents the following challenges:
a) Variations in handset microphones which result in
severe mismatches between speech data gathered
from these microphones.

b) Signal distortions due to the telephone channel.

c) Inadequate control over speaker/speaking
conditions.

Consequently, speaker identification and verification
systems have not yet reached acceptable levels of
performance over the telephone network. Several
feature extraction techniques are explored but only th
Power Spectral Densities (PSDs) based technique is
reported in this paper. The discrete Fourier transform of
the telephone voice samples is obtained and the PSDs
are computed. The PSDs of three different speakers A,
B and C uttering the same phrase is shown in figures 3,
4 and 5 respectively.

0 1000 2000 3000 4000 5000 6000 7000 8000
-80
-60
-40
-20
0
Power Spectrum Magnitude (dB)
Frequency Hz
Figure 3: PSD of Speaker A
0 1000 2000 3000 4000 5000 6000 7000 8000
-100
-80
-60
-40
-20
0
Power Spectrum Magnitude (dB)
Frequency Hz
Figure 4: PSD of Speaker B
0 1000 2000 3000 4000 5000 6000 7000 8000
-150
-100
-50
0
Power Spectrum Magnitude (dB)
Frequency Hz
Figure 5: PSD of Speaker C
It can be seen from these figures that the PSDs of th
speakers differ from each other. Ref. 5 has reported
success on speaker identification up to 66% and 90%
with PSDs as input vectors to multilayer feedforward
neural networks and Self-Organizing Maps ( SOMs)
respectively.

4. PATTERN MATCHING USING ARTIFICIAL
NEURAL NETWORKS
Artificial Neural Networks (ANNs) are intelligent
systems that are related in some way to a simplified
biological model of the human brain. They are
composed of many simple elements, called neurons,
operating in parallel and connected to each other by
some multipliers called the connection weights or
strengths. Neural networks are trained by adjusting
values of these connection weights between the
neurons.

Neural networks have a self learning capability, are
fault tolerant and noise immune, and have applications
in system identification, pattern recognition,
classification, speech recognition, image processing,
etc. In this paper, ANNs are used for pattern matching.

The performance of different neural networ
architectures are investigated for this application. Thi
paper presents results for the MLP feedforward network
and the self-organizing feature map. Descriptions of
these networks are given below.

4.1. MLP FEEDFORWARD NETWORK
A three layer feedforward neural network with a
sigmoidal hidden layer followed by a linear output laye
is used in this application for pattern matching. The
neural network is trained using the conventional
backpropagation algorithm. In this application, an
adaptive learning rate is used; that is, the learning rate is
adjusted during the training to enhance faster global
convergence. Also, a momentum term is used in the
backpropagation algorithm to achieve a faster global
convergence.

The MLP network in figure 6 is constructed in the
MATLAB environment 6. The input to the MLP
network is a vector containing the PSDs. The hidden
layer consists of thirty neurons for four speakers. The
number of neurons in the output layer depends on the
number of speakers and in this paper it is four.

sigmoidal activation function
linear activation function
1st speaker
Nth speaker
Vector
of PSDs
Figure 6: MLP Network
An initial learning rate, an allowable error and the
maximum number of training cycles/epochs are the
parameters that are specified during the training phase
to the MATLAB neural network program.

4.2. SELF-ORGANIZING FEATURE MAPS
The second type of neural network selected for this
investigation is the self-organizing feature map 7. This
neural network is selected because of its ability to learn
a topological mapping of an input data space into a
pattern space that defines discrimination or decision
surfaces. The operation of this network resembles the
classical vector-quantization method called the k-means
clustering. Self-organizing feature maps are more
general because topologically close nodes are sensitive
to inputs that are physically similar. Output nodes will
be ordered in a natural manner.

Typically, the Kohonen feature map consists of a two
dimensional array of linear neurons. During the training
phase the same pattern is presented to the inputs of each
neuron, the neuron with the greatest output value is
selected as the winner, and its weights are updated
according to the following rule:
w t w t x t w t i i i ( ) () () () + = + ;#8722; 1 a (1)
where wi(t) is the weight vector of neuron i at time t,
is the learning rate and x(t) is the training vector.

Those neurons within a given distance, the
neighborhood, of the winning neuron also have their
weights adjusted according to the same rule. This
procedure is repeated for each pattern in the training set
to complete a training cycle or an epoch. The size of the
neighborhood is reduced as the training progresses. In
this way the network generates over many cycles an
ordered map of the input space, neurons tending to
cluster together where input vectors are clustered,
similar input patterns tending to excite neurons in
similar areas of the network.

5. IMPLEMENTATION OF THE SPEAKE
IDENTIFICATION AND VERIFICATION SYSTEM
The work that is being reported in this paper is
implemented in software. The telephone speech i
captured and processed on a Pentium II 233 MHz
computer with a 16 bit sound card. The telephone
receiver is interfaced to the sound card. Telephon
speech is captured over signals transmitted within 10
kilometres of transmission network. Digital signal
processing and neural network implementations are
carried out using the MATLAB signal processing and
neural network toolboxes respectively. This work is
currently undergoing and an implementation of a realtime
speaker identification and verification system ove
telephone lines on a digital signal processor is
envisaged.

6. EXPERIMENTAL RESULTS
The MLP network is trained with the PSDs of eight
voice samples recorded at different instants of time
under controlled and uncontrolled speaking conditions
of four different speakers uttering the same phrase at all
times. Controlled speaking conditions refer to noise and
distortion free conditions unlike uncontrolled speaking
conditions which have noise and distortion on the
transmission lines. The number of PSD points for each
voice sample is about 500. As mentioned in section 4.1,
an adaptive learning rate is used for the MLP network.

The initial learning rate is 0.01. The allowable sum
squared error and maximum number of epochs
specified to the MATLAB neural network program i
0.01 and 10000 respectively. It is found that the sum
squared error goal is reached within 1000 epochs.

A success rate of 100% is achieved when the trained
MLP network is tested with the same samples used in
the training phase. However, when untrained samples
are used, only a 63% success rate is obtained. This is
due to the inconsistency in the PSDs of the input
samples with those used in the training phase. The MLP
network is also tested with unseen voice samples of
people who are not included in the training set and the
network successfully classified these voice samples as
unidentified.

Four speakers are identified using the self-organizing
feature map like in the case of the MLP network. An
initial learning rate of 0.01, an allowable sum squared
error of 0.01 and a maximum of 70000 epochs are
specified at the start of the training process to the
MATLAB neural network program. The results with the
self-organizing feature map shows a drastic change in
the success rate in identifying the speakers as reported
in 5. With PSDs as inputs, a success rate of 85% and
90% is achieved under uncontrolled and controlled
speaking conditions respectively.

Ref.5 has reported that success rate can be increased
to 98% under uncontrolled speaking conditions by
using Linear Prediction Coefficients (LPCs) as inputs to
SOMs which remains to be yet to be tried out in this
work. Currently, with the PSDs as inputs a lot of
computations is involved and the SOM takes a lot of
time to learn.

7. CONCLUSIONS
This paper has reported on the feasibility of using
neural networks for speaker identification and
verification over short distance telephone lines and ha
shown that performance with the self-organizing map is
higher compared to that with the multilayer feedforward
neural network. Different feature inputs to the selforganizing
map remains to be tried out in order to
achieve higher identification/verification rates
minimizing the training time and the size of the
network. Speaker identification with telephone speech
signals over long distance telephone lines is currentl
being investigated using similar techniques.

This paper has shown that speaker identification is
possible over the telephone lines and therefore
telephonic bank and other transactions can be
authenticated. Hence a technique to combat and/or
reduce white collar crime.

8. REFERENCES:
1 D.A.Reynolds, “Large population speake
identification using clean and telephone speech”, IEEE
Signal Processing Letters, vol. 2 no. 3 March 1995, pp.

46 – 48.

2 J.M.Naik, L.P.Netsch, G.R.Doddington, “Speaker
verification over long distance telephone lines”,
Proceedings of IEEE International Conference on
Acoustics, Speech, and Signal Processing (ICASSP),
23-26 May 1989, pp. 524 – 527.

3 A.L.Mcilraith, H.C.Card, “Birdsong Recognition
Using Backpropagation and Multivariate Statistics”,
Proceedings of IEEE Trans on Signal Processing, vol.

45, no. 11, November 1997.

4 G.K.Venayagamoorthy, V.Moonasar,
K.Sandrasegaran, “Voice Recognition Using Neural
Networks”, Proceedings of IEEE South African
Symposium on Communications and Signal Processing
(COMSIG 98), 7-8 September 1998, pp. 29 – 32.

5 V.Moonasar, G.K.Venayagamoorthy, “Speaker
identification using a combination of different
parameters as feature inputs to an artificial neural
network classifier”, accepted for publication in the
Proceedings of IEEE Africon 99 conference, Cape
Town, 29 September – 2 October 99.

6 H.Demuth, M.Beale, MATLAB Neural Network
Toolbox User’s Guide, The Maths Works Inc., 1996.

7 T.Kohonen, Self-organizing and associate memory
Spring Verlag, Berlin, third edition, 1989.