IRJET- Hand Talk- Assistant Technology for Deaf and Dumb

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 06 Issue: 07 | July 2019 www.irjet.net p-ISSN: 2395-0072
© 2019, IRJET | Impact Factor value: 7.211 | ISO 9001:2008 Certified Journal | Page 3555
Hand Talk- Assistant Technology for Deaf and Dumb
Varsha K1, Hamsini C2
Students, Dept. of Electronics and Communication, The National Institute of Engineering, Mysuru, Karnataka,
India
---------------------------------------------------------------------***----------------------------------------------------------------------
Abstract – In this paper we represent smart glove for
deaf and dumb patient. About nine billion people in the
world are deafanddumb.Thecommunicationbetweena
deaf normal visual people. This creates averylittleroom
for them with communication being a fundamental
aspect of human life. The blind people can talk freely by
means of normal language whereas the deaf-dumbhave
their own manual-visual language known as sign
language. Sign language is a non-verbal form of
intercourse which is found amongst deafcommunitiesin
world. The languages do not have a common origin and
hence difficult to interpret. The project aims to facilitate
people by means of a glove-based communication
interpreter system. The glove is internallyequippedwith
five flex sensors. For each specific gesture, the flexsensor
produces a proportional change in resistance. The
processing of these hand gestures is in Arduino-Uno
Board which isanadvanceversionofthemicrocontroller
and the LABVIEW software. It compares the inputsignal
with predefined voltage levels stored in memory.
According to that required sound is produced which is
stored is memory with the help of speaker. In such a way
it is easy for deaf anddumbtocommunicatewithnormal
people.
Keywords – Deaf and Dumb Patient, Sign Language,
Flex Sensor
1. INTRODUCTION
“Communication Is the Fundamental Human Right”.
Communication is a two-way interactiveprocessandit
plays a major role in daily life. Communication
strategies can affect one’s ability to engage in
interaction and cope up with social conflict, but this
communication is not possible for the people who are
deaf and dumb. About nine billion people at intervals
the planet unit of measurement dumb.
The main aim of this project is to build a bridge
between them by developing a portable
communication intheformofgestureglove.Thegloves
helpininterpretingthesignlanguagethroughgestures.
Languageisalsoanon-verbalformofintercoursethat's
found among deaf communities.Thelanguageshaven't
got a typical origin and hence hard to interpret. A
Dumb communication interpreter is also a tool that
interprets the hand gestures to sensibility speech.
Gestures play a major role in the daily activities of
human life, during communication providing easier
understanding. In other words, Gesture recognition
refers to recognizingmeaningfulexpressionsofmotion
by a human, involving the hands. Between all the
gestures performed, hand gestures play an important
role which helps us to express more in less time. For
each hand gesture created, a symptomis formedbythe
sensors appreciatethegestureandthemicrocontroller
matches the gesture withpre-storedinputs.Thedevice
exclusively interprets alphabets, numbers and few
phrases of American sign language. A training mode is
gettable on the device therefore it fits every user and
accuracy is inflated.
The Gesture glove is a normal, clothdrivingglovefitted
with flex sensors along the length of each finger and
the thumb. The sensors output a stream of data that
varies with degree of bend. The outputfromthesensor
is analog values it is converted to digital andprocessed
by using microcontroller and then it will be
transmitted wirelessly through Bluetooth module
HC05 to the android mobile application where the
received data will be displayed and converted to
speech where the audio can be heard
Sign Language is the means of communication among
the deaf and mute community. Sign Language emerges
and evolves naturally within hearing impaired
community. Sign Language communication involves
manual and non-manual signals where manual signs
involve fingers, hands, arms and non-manual signs
involve face, head, eyes and body. Sign language is a
complete natural language that uses different ways of
expressionforcommunicationineverydaylife.Theaim
of the sign language recognitionsystemistopresentan
efficient and accurate mechanism to transcribe text or
speech, thus the “dialog communication” between the
deaf and hearing person will be smooth. There is no
standardized sign language for all deaf people across
the word. A person who can talk and hear properly
(normal person) cannot communicate with deaf &

dumb person unless he/she is familiar with sign
language. Same case is applicable when a deaf & dumb
person wants to communicate withanormalpersonor
blind person.
ASL is a language completely separate and distinct
from English.Itcontainsallthefundamentalfeaturesof
language—it has its own rules for pronunciation,word
order, and complex grammar. While every language
has waysofsignalingdifferentfunctions,suchasasking
a question rather than making a statement, languages
differ in how this is done. For example, English
speakers ask a question by raising the pitch of their
voice. ASL’s has all of the features of any language; that
is, it is a rule- governed system using symbols to
represent meaning. In ASL, the symbols are specific
hand movements andconfigurations that aremodified
by facial expressions to convey meaning. These
gestures or symbols are called signs. It is a unique and
distinct language, one that does not depend on speech
or sound. ASL has its own grammar, sentence
construction, idiomatic usage,slang,style,andregional
variations-the characteristicsthatdefineanylanguage.
American Sign Language is the shared language that
unites Deaf people in what is known as the Deaf
community.
Fig -1: ASL hand gestures for alphabets
Fig-2: ASL hand gestures for numbers
Fig-3: ASL hand gestures for emotions
2. LITERATURE SURVEY
[1] A group of engineering studentsatCarnegie Mellon
University, Bhargav Bhat, Hemant Sikaria, Jorge L.
Meza andWesleyJindemonstratedtheirproject „Hand
Talk‟asensorequippedglovethattranslatesfinger
and hand gestures into spoken words. This is the
first demonstrator model to show the functionality
based on a limited vocabulary of 32 words. Sensors in
the glove pick up gestures and transmit the data
wirelessly via Bluetooth to a cell phone which runs
Text to Speech software.Thesensordataareconverted
first into text and then to voice output. A person not
knowledgeable in Sign language can listen via the cell
phone what the other person is saying inSignlanguage
form. The main advantage with this design was its
simplicity and the cheap components these students
used to create this amazing and truly interactive glove
that could help to improve greatly the communication
barrier between deaf persons and people. The latest
sensor being used for the Hand Talk glove is the
accelerometer. Instead of workingintwoplanes(Xand
Y) like in the flex sensors, it works in X-Y, Y-Z and X-Z
planes. It is more reliablethantheflexsensorsandonly
one accelerometer is required for one glove. More
number of programs can be fed into it so it can
accommodate more number of sounds in it.
[2] Hand Gesture Recognition System For Dumb
People: Authors presented the static hand gesture
recognitionsystem usingdigital imageprocessing. For
hand gesture feature vector SIFT algorithm is used.
The SIFT features have been computed at the edges
which are invariant to scaling, rotation, addition of
noise.
[3] An Automated System for Indian Sign Language
Recognition in: In this paper a method for automatic

recognition of signs on the basis of shape based
features is presented. For segmentation ofhandregion
from the images, Otsu’s thresholdingalgorithmisused,
that chooses an optimal threshold to minimize the
within-class variance of thresholder black and white
pixels. Features of segmented hand region are
calculated using Hu’s invariantmomentsthatarefedto
Artificial Neural Network for classification.
Performance of the system is evaluated on the basis of
Accuracy, Sensitivity and Specificity.
[4] Glove-based systems: Glove-system is composed
of an array of sensors, electronics for data acquisition
or processing ,power supply and a support for the
sensors that can be worn on the user’s hand. Sayre
glove, LED glove, data gloves (used here) and
cyberloves are the different types of gloves used.
Gesture- based applications of glove systems involves
extraction and classification of features to
automatically understand gestural languages used by
the deaf community.
[5] Glove Talk Neural networks: were used to
implement an adaptive Digital Electronics interface,
called Glove Talk, which maps hand gesturestocontrol
the parameters of a parallel formant speech
synthesizer to allow a user to speakwithhishands.Itis
used to implement an artificial vocal tract.
[6] Real-time gesture recognition: A prototype
system has been implemented with a lexicon of 250
vocabularies in Taiwanese Sign Language (TWL). On
detecting the beginning of posture holding, the system
extracts features, including position, orientation, and
posture, and also starts tracking motion trajectory.
[7] American sign-language interpreter: When the
user makes a hand gesture, the binary values offingers
were checked for 5 times and if they matched, then the
microcontroller indicated the gesture as valid and
perfect match and displayed the corresponding codes
on LCD.
[8] Gesture-deployed system: We present our
prototype for a gesture recognizing glove (data glove).
We use low cost packaging material for making
piezoresistivesensors.Theseflexsensorsdetectabend
in fingers and we map this data to a character set by
implementing a Minimum Mean SquareErrormachine
learning algorithm. The recognized character is
transmitted via Bluetooth, to an Android phone, which
performs a text to speech conversion. Our motivation
for Hand Talk is to compare hand configurations with
sign language charts and generate artificial speech
which articulates the gestured words. This technology
also has further applications as a 3D mouse, virtual
keyboard, control for precision control of roboticarms
[9] Sign Language Translator Gloves: Hand motion
tracking and gesture recognition are a fundamental
technology in the field of proactive computing for a
better human computer interaction system. In this
paper, we have developed a 3-D hand motion tracking
and gesture recognition system via a data glove
(namely the KHU-1 data glove consisting of three tri-
axis accelerometer sensors, one controller, and one
Bluetooth). The KHU-1 data glove is capable of
transmitting hand motion signals to a PC through
wireless communication via Bluetooth. Also we have
implemented a 3-D digitalhandmodelforhandmotion
tracking and recognition. The implemented 3-D digital
hand model is based on the kinematic chain theory
utilizing ellipsoidsandjoints.Finally,wehaveutilizeda
rule-based algorithm to recognize simple hand
gestures namely scissor, rock, and paper using the 3-D
digital hand model and the KHU-1 data glove. Some
preliminary experimental results are presented in this
paper
[10] Mister Gloves: Gesture recognition system is a
system to interpret movement of hand or head via
algorithms. Algorithms to interpret gesture are in the
form of software, hardware or combination of both.
The main goal of gesture recognition system is to
enable humans to communicate with machine. This
paper present the development of wireless Bluetooth
hand gesture recognition system using six 3-axis
accelerometers embedded in a glove and a database
system in a computer. This system can recognize any
sampled data saved in the database while promoting
maximum portability and mobility to the user via
wireless Bluetooth technology. Analyses such as static
data, dynamic data, and average recognition rates
relationships are discussed in this paper
[11]. Smart Glove Project: Dumb persons
communicate through gestures which are not
understood by the majority of people. Gesture is a
movement of part of the body, especially a hand or the
head, to express an idea or meaning. This paper
proposes a system that converts gestures given by the
user in the form of English alphabets into
corresponding voice and translates this English voice
output into any other Microsoft supported languages.
The system consists of MPU6050 for sensing gesture
movement, Raspberry pi for processing, three button

Keypad and speaker. It is implemented by using
trajectory recognition algorithm for recognizing
alphabets. Raspberry pi generates voice output for the
text in multiple languages using voice RSS and
Microsoft translator. When tested, the system
recognized A-Z alphabets and generated voice output
based on the gestures in multiple languages.
[12]. Magic Glove : The primary objective of the paper
is to construct and test a low-cost, minimally
supervisedgesturerecognitionsystemwhichidentifies
static gestures efficientlyandaccurately.Theproposed
system uses ADXL335 accelerometer sensors which
track the gestures and these sensors are interfaced
with an Arduino ATMega 2560 micro-controller for
data processing and gesture recognition. The software
of the system implemented in the micro-controller,
features a computationally feasible algorithm which
requires only nominal resources to recognize the
gestures. The paper further elucidates on minimizing
the number of accelerometers to reduce the cost and
power-consumptionofthesystem.Theperformanceof
the system is assessed using static gestures in the
alphabets of the American Sign Language (ASL) across
data-sets obtained from 3 trained ASL signers
3. METHODOLOGY
The gesture gloves should be worn by deaf and
dumb people. Gloves is attached with five flex sensors.
For every gesture in sign language there is change in
finger position and bending. This change or bending of
the fingers corresponds to change in resistance of flex
sensors. This data from flex sensors is processed in
Arduino UNO microcontroller. The flex sensor used is
Arduino UNO compatible and it is connected to
Arduino UNO through wires. Arduino UNO is powered
with power module which supply 5V.
This data is wirelessly transmitted by Bluetooth
module (HC05), present on thegestureglove.Thisdata
is received by the Bluetooth module of mobile. The
android mobile application is designed such that it
receives the processed data throughtheblue toothand
displays the received data on the mobile screen and
also simultaneously converts the text received into
speech and the audio can be heard. The output
obtained is both display and audio of the
corresponding gesturemadebythedifferentiallyabled
people.
Fig-4: Block diagram for left hand microcontroller
operation
Fig-5: Block diagram for right hand microcontroller
operation
3.1 Description of left hand flow of the project
Each sign in ASL has particular gesture. Five flex
sensors are attached to gesture glove to identify the
change in resistance value of the flex sensor when
bending of fingers is encountered. Whenever there is
change in the resistance of the flex sensors is
encountered, the resistance converted voltage value
which is seen as the output of flex sensors will be
received by the Arduino UNO and will be sending the
same to the right hand Arduino UNO serially through
serial port. If there is no change in resistance value if
flex sensors encountered, then the wait condition
comes into picture.
Fig-6: Work flow of left hand processing unit

3.2 Description of right hand flow of the project
When the system is turned on, the conditionwhether
the Arduino UNO is in GROUND mode and 5V mode is
checked. There are two modes which we use in this
project
GROUND mode: both Left and Right hands are in
operation.5V mode: only Right hand is in operation. If
the GROUND mode is on, then the right hand Arduino
UNO receives data from left handArduinoUNOtooand
the same processing is done.
If the 5V mode is on, then only right hand flex sensors
data is received. The processed data is then
transmitted to mobile application which displays the
data received and also converts and received data to
speech where the corresponding audio is heard.
Fig-7: Work flow of right hand processing unit
3.3 Arduino Code Description
1.Firstly the right hand code.
2.We declare the variables for all the 10 fingers.
3.We declare a variable for mode and initialized as
zero.
4.We declare variable for string input and initialized it
as zero.
5.We declare a variable forstringoutputandinitialized
it to null.
6.We define a function called split where it accepts the
string value until the in value count is specified where
the count is increasedbyonewheneveritcomesacross
a “,” and finally returns the complete string.
7.The next part of the code runs in the loop where the
analog input from the flex sensors are read
continuously.
But there are two modes to be considered which are
high and low.
8.When the mode is high, it’s the phrase mode and the
number mode where it accepts 10 inputs until there is
a next line command.
9.The function that converts the strings accepted into
integers which has to be displayed on the serial
monitor. If the integers match the range of values for a
particular phrase or number, then it displays the same
on the serial monitor or it prints the values as it is.
10 .If the mode is low, then it’s the alphabet mode. It
only accepts the 5 inputs and search foranymatchesin
the range of values assigned for the differentalphabets
and prints the same in the serial monitor. Otherwise it
displays just the values.
11. Lastly the left hand Arduino code, As the complete
processing is done in the right hand controller, the
work of left hand controller is to just accept the 5
inputs from the flex sensors and display it on the serial
monitor.
Fig-8: Flow chart of Arduino code
3.4 System Algorithm
Arduino IDE is used for coding in the proposedsystem.
The syntax and the statements in Arduino IDE is
similar to C syntax. Hence itiseasiertounderstandand
debug the code. Some of the important algorithms and
syntaxes are discussed below.

The steps in the algorithm are as follows:
Variables Initialization: The flex sensors attached to
each finger is assigned with the variable. The gestures
are trained based on the values attained by these
variables.
Mode setup: The two modes are defined using the
switch operation.Theyarealphabetmodeand,number
and phrase mode.
Pattern defining: Each pattern is defined first with
respect to the values taken from the flex sensors in
accordance with the American Sign Language and
pattern matching algorithm is used.
Pattern matching: The gloves are trained with some of
the gestures according to the American Sign Language.
To the corresponding values of thegesture,thegesture
is identified using the pattern matching algorithm.
EXPERIMENTAL RESULTS
The project developed focuses mainly to achieve the
day-to-daycommunicationamongthenormaland,deaf
and mute people in an efficient and simple way, as the
normal people cannot understand the American Sign
Language which is used by the challenged people.
The Gesture gloves project has successfullytrainedthe
gloves with 26 alphabets, 9 numbers and frequently
used phrases. The gloves employ the gestures using
both the hands which has made the project unique and
is more reliable than the existing projects. The
Bluetooth module used make the system portable as it
is wireless.
Mobiles have become the essential necessity for being.
Hence, the application being developed using android
studio displays the message and speech is obtained.
This eases the process of understanding.
The ambiguity is very less when compared to the
gesture gloves project employed using accelerometer
as the flex sensors give out the apt values and is easyto
train.
Test results show that the recognition of the gesture
with accuracy than the existing systems.
Future implementations would be employing the
gestures being trained with the prediction of the
further conversation using the suitable modules and
algorithm for easy communication of deaf and mute
people with the normal people.
CONCLUSIONS
American Sign Language (ASL) is a complete, complex
languagethatemployssignsmadebymovingthehands
combined with facial expressions and postures of the
body. The gesture gloves project which is trained
according to the American sign language. Deaf people
rely on sign language interpreters for communication.
However, they cannot depend on interpreters in
everyday
life mainly due to high costs and difficulty in finding and
scheduling qualified interpreters. Hence,thisproject enables
communication with deaf and is one of several
communication options used.
In order to express the wanting, or in the case of emergency
or to express the feelings, communication playsa majorrole,
this project is a useful tool for speech impaired and partially
paralyzed patients which fill the communication gap
between patients, doctors and relatives.
The gestures in the ASL will mostly use two hands. This
project helps the deaf and speech impaired to sign their
expression and enable the communicationwiththecommon
people. The dumb gets a voice to express himself.AsArduino
Uno is used instead of other bulk microcontroller or bulk
electronic circuit, which makes it user friendly and compact
and satisfies our memory requirement to the project.
Arduino Uno is most compatible with flex sensors.
Thus the project is a useful and fully functional real-world
product that efficiently translates the movement of the
fingers for the fingerspelling of American Sign Language
(ASL).
Use of Bluetooth module eliminates the bulk circuit at the
receiver. Using an android application which receives the
text from the Bluetooth module connected to Arduino
displays the text and converts it to speech. The application
model becomes more compact and efficient as it developed
as a simple user friendly application.
REFERENCES
[1] S. Siavoshi, "Deaf education in Iran, problems and
strategies," The Iranian deaf studies foundation, 2010
[2] K.C. Shriharipriya and K. Arthy, “Flex sensor based
hand gesture recognition system”, in Proceedings of
International Journal Of Innovative Research And
Studies (IJIRS), Vellore, India, May 2013
[3] AjinkyaRaut, VineetaSingh,Vikrant Rajput,
RuchikaMahale, “HAND SIGN INTERPRETER”, in
Proceedings of The International Journal ofEngineering
And Science (IJES), Pune, India, December 2012

[4] Bhavina Patel, Vandana Shah, RavindraKshirsagar,
“MICROCONTROLLER BASED GESTURE RECOGNITION
SYSTEM FOR THE HANDICAP PEOPLE”, in Proceedings
of Journal of Engineering Research and Studies, Surat,
India, December 2011

IRJET- Hand Talk- Assistant Technology for Deaf and Dumb

More Related Content

What's hot (20)

Similar to IRJET- Hand Talk- Assistant Technology for Deaf and Dumb (20)

More from IRJET Journal (20)

Recently uploaded (20)

IRJET- Hand Talk- Assistant Technology for Deaf and Dumb