International Conference on Advances in Computing, Communication Control and Networking (ICACCCN2018)

Voice Recognition Application Based Home

Automation System with People Counter
Gudipati Sravanthi, Gottipati Madhuri
EEE Department,Vignan Institute of technology and Science,
Deshmukhi, Yadadri District, Telangana, India.
Email: sravanthi244.g@gmail.com, gottipati.madhuri@gmail.com

Nipun Sharma, Ashu Tiwari, Abhishek Kashyap, B. Suresh

ECE Department, Jaypee Institute of Information Technology,
Noida, Uttar Pradesh, India.
Email: nipunsharma33@gmail.com, ashu.tiwari945@gmail.com,
abhishek.kashyap@jiit.ac.in, b.suresh@jiit.ac.in

Abstract—Home Automation is a new technology that latest android technology and along with the help
is growing in this era. It has the capability to provide of blue-tooth facilities in a Mobile provides the
supporting systems to the fast changing world, elderly and feature of giving commands and then processing it
physically challenged people. This paper explains the use of
into logics which help in switching ON/OFF the
overall design which has been built and implemented. The
Wireless Home Automation System (WHAS) with the help loads. [11]
of Voice Recognition Application, makes the installation Drawback here is that the systems are not robust
and implementation, cheaper and easy to use.Foot Step when voice commands [3] are given from a far
Counter is the second part of this paper which deals with distance. Other drawback is, it does not give the
the automatic switching of lights. Its need is, when there current status of the appliance as it is switched
is no time to start the application, or to connect through
ON/OFF. In that case after giving voice commands
Blue-tooth, the person passes door and lights within the
room will be switched.In this paper, we have observed only we are able to know the status.So to overcome
range of voice recognition application to be 1.5 meters. this issue, we use People Counter, which is the
The voice recognize accuracy achieved is 85.25% with four combination of sensors. This way we can make the
parameters of access groups. Home Automation System more efficient. [12]
Index Terms—Home Automation, WHAS, smart home,
GSM
II. S YSTEM OVERVIEW
I. I NTRODUCTION The Home Automation System (HAS) connects
As the world is advancing, automation is all the electronic devices, appliances, gadgets and
spreading everywhere even to our homes. Regarding loads with homework setup. [13] Home Automation
this, a new automation system is designed which system work with the help of different electronic
not only saves time but also helps the elderly and modules to form an integrated system. This system
disabled people in making their life easy. [1] The includes Smart-phone with Voice Recognition App,
design basically focuses on switching ON/OFF Blue-tooth module, Relay module, Arduino uno and
various loads in the house through voice. This load appliances as shown in Figure 1. The app is
technology provides long term performance which developed by MIT App Inventor. [9] The features
will support the homes with comfort. [2] The included in the app are Speech Recognizer, Blue-
design uses the blue-tooth technology and Android tooth Client and clock. Smart-phone is a cheap and
platform to provide the facility of home automation. less expensive wireless remote control for HAS.
An application is build which interacts with the Foot Step Counter is the technique which is an

ISBN: 978-1-5386-4119-4/18/$31.00 ©2018 IEEE 574

 International Conference on Advances in Computing, Communication Control and Networking (ICACCCN2018)

Fig. 2. Unit of the People Counter

Fig. 1. Functional block diagram of the Wireless Home Automation

(WHAS)

addition to the HAS. It needs sensors to detect

and a counter which keeps the count of persons
passing through the door or any gate. This saves
the resources and time for process.
The voice is transferred to the recognition app
through microphone of the smart-phone. [4] It con-
verts the voice in terms of the string. At starting
Fig. 3. People Counter based Smart home
of the string it adds the “*” which indicates start
of string. Then it is terminated with “#” indicating
end of the string. After string is reformed, it is
transferred bit-by-bit to the Blue-tooth module. Bits sensors. This is done to store the count of people
are then accumulated in the micro-controller (e.g.:- going in and out of the room. It is a technique which
8051, Arduino uno, AVR). Now by assigning a all the practical systems like security system [10] in
particular word or a group of words for appliances offices,etc. are using nowadays. [15] But, the system
within a room(e.g. *Bedroom on#,*Bedroom off#, fails to implement in the daytime, as we do not want
for switching ON/OFF bulb) an appliance can be lights ON in daytime. So, in external circuit we use
controlled. a timer. This timer can be used for a long time,
The People Counter on the other hand is the com- i.e. up to 100 years. This makes the system reliable
bination of sensors which detects motion on the and also it takes power from a cell that is installed
passage (e.g. a door, main gate, industry passage, in the circuit externally. Timer helps the sensors in
etc.) keeping in mind that people can go in and out knowing in which time to operate. So, generally the
of the door simultaneously sensor helps in storing timer will deactivate the sensors power supply in the
the count of people as shown in Figure 2. It detects daytime. This will make the system more robust and
the motion with help of IR sensors, [8] and sends the less power consuming.
data to the micro-controller. The pin gets high(+5V) There are more add-ons which can make the
with respect to that sensor pin which was connected HAS more robust and efficient. Likewise, GSM
to the micro-controller. Hence, the load is switched module can be used for initiating call and message
ON. The people counter is made in pair of two on cell phones by the help of these sensors. This

575
 International Conference on Advances in Computing, Communication Control and Networking (ICACCCN2018)

will make the home more secure. At last, there is

a lot of advancement in this field. That is the HAS
can be controlled from a remote area of the globe.

This will make the corners of the world intact

with this technology without the human interven-
tion.
Figure 3 shows two sensors(color-skin) plugged
on the door(color-blue) in home. The counter shows
when a person passes from one room to other, the
IR sensor detects the person and the lights of that
room glows. This computation is done by micro-
controller(arduino uno).
III. O BSERVATIONS
The HAS was tested on the basis of certain Fig. 4. Mean parameters for access group [6]
parameters which includes the effective detection of
voice commands and range till which app responses
accurately. Figure 4 shows various access groups.
We have taken 4 voice commands. These are based
on the need of residents. As the voice commands
are recognized, [5] the respective actions are send
to the micro-controller(Arduino uno) at a baud rate
of 9600 bps. These groups are subdivided into ON
and OFF. 100 attempts were made for each access
group, which was subdivided into 50 for ON and
50 for OFF.
• The 1st access group controls the lights of
Bedroom.It consist of two voice commands:-
BEDROOM ON/OFF.Number of successful at-
tempts for BEDROOM is 90/100.
• The 2nd access group controls the lights of
Bathroom.It consist of two voice commands:-
BATHROOM ON/OFF.Number of successful Fig. 5. Accuracy of the voice recognition app
attempts for BATHROOM is 88/100.
• The 3rd access group controls the lights of
Kitchen.It consist of two voice commands:- voice commands. This is an essential observation,
KITCHEN ON/OFF.Number of successful at- as it also tells that after noise is added from the
tempts for KITCHEN is 80/100. environment voice is recognized properly by app or
• The 4th access group controls the lights not. We can see from the Figure 5, as the distance
of Drawing room.It consist of two increases due to addition of noise the app efficiency
voice commands:- DRAWING ROOM decreases and at last the app stopped responding.
ON/OFF.Number of successful attempts for The app was tested by voice command ”HELLO”.
DRAWING ROOM is 79/100. It was tested for 5 times from a particular distance.
These access groups can also include appli- Then it was tested for far distances.
ances other than the light eg. fans, cooler, A.C’s, • It was observed that the app was working with
geyser,etc. Figure 5 shows the efficiency in terms of full efficiency till 0.5 meters. After 0.5 meter
range, that how much far the app is recognizing the performance decreased.

576
 International Conference on Advances in Computing, Communication Control and Networking (ICACCCN2018)

Fig. 6. Voice Recognition App home page(left) and Bluetooth client Fig. 7. Prototype-Smart Home Automation System
list-name and addresses(right)

• At 1.0 meters the app worked 3/5 times cor-

rectly.
• At 1.5 meter the app worked 1/5 times cor-
rectly.
• At 2.0 meter the app stopped responding.

By these observations it is concluded that the app’s

range is till 1.5 meters. But, the app will remain
connect to appliances till 20 meters, as blue-tooth
is used. [14]

IV. R ESULTS
The home automation system helps in monitoring
and controlling the appliances.Android App has
been made by the help of MIT App Inventor and Fig. 8. Prototype - People Counter
prototype has been tested. Figure 6 shows the Voice
Recognition App. Left screen is the home page
consisting of microphone button. After connecting Figure 8 shows the hardware for People Counter.
app via blue-tooth it activates. The app then gets In this system we have 2 rooms named Bedroom
connected with blue-tooth module(HC-05) on the and Kitchen. It works with IR sensors which are
system. Now on pressing microphone button in the installed back and front on the doors. These sensors
app a clock starts and accepts voice commands detect motion in front of them.These sensors send
and converts into string. Then this string is sent signal to micro-controller, and it converts the signal
to the micro-controller bit-by-bit via blue-tooth. into bits and with the help of relay the lights are
Figure 7 shows the prototype’s experimental setup controlled. LED Display shows the real time status
for Voice home automation system with 4 rooms of the room. It shows the number of persons in
each installed with a light bulb. The system consists the room. When a person enters the door, counter
of micro-controller(Arduino uno), relay, and blue- number increases. The counter is set by micro-
tooth module. controller and the status is checked every time. A

577
 International Conference on Advances in Computing, Communication Control and Networking (ICACCCN2018)

door has two IR sensors. The sensor outside room of energy. Further with the help of Internet Con-
counts number of person entering in the room i.e. nectivity one can check the status of appliances and
increments the counter and other sensor which is control them from distant place.
installed inside the room detects people in the room R EFERENCES
and thus decrements the counter. Home Automation
[1] Humaid AlShueili, Gourab Sen Gupta, Subhas Mukhopadhyay,
System along with People Counter forms an inte- “Voice Recognition Based Wireless Home Automation System,”
grated system which not only automates the home in 4th International Conference on Mechatronics (ICOM), Kuala
but also makes the home smart, improves security, Lumpur, Malaysia,17-19 May 2011.
[2] B. Lecouteux, M. Vacher and F. Portet, “Distant speech recog-
energy,robustness and becomes easy to use. Further nition for home automation: Preliminary experimental results in a
with the help of GSM module or Wifi module we smart home,” in 2011 6th Conference on Speech Technology and
can control this type of systems from a distant place Human-Computer Dialogue (SpeD),Brasov, 2011, pp. 1-10.
[3] K. Gnanasekar, P. Jayavelu and V. Nagarajan, “Speech recogni-
via Internet, also known as Internet Of Things(IOT). tion based wireless automation of home loads with fault identifica-
By this we can also build Smart Factories which is tion for physically challenged,” in 2012 International Conference
based on Industry 4.0. on Communication and Signal Processing,Chennai, 2012,pp. 128-
132.
[4] M. Tharaniya soundhari and S. Brilly Sangeetha, “Intelligent
V. C ONCLUSION interface based speech recognition for home automation using
android application,” 2015 International Conference on Inno-
The home automation system with voice recog- vations in Information, Embedded and Communication Systems
nition app was built and implemented successfully. (ICIIECS), Coimbatore,2015,pp. 1-11.
[5] Minh-Son NGUYEN, Tu-Lanh VO, “Vietnamese Voice Recog-
While using voice commands care must be taken nition for Home Automation using MFCC and DTW Techniques,”
by using system in noise free environment. The International Conference on Advanced Computing and Applica-
working range of the system has been improved. tions(ACOMP),Ho Chi Minh City,2015,pp. 150-156.
[6] Y. Mittal, P. Toshniwal, S. Sharma, D. Singhal, R. Gupta and
The system responds efficiently till 1.5 meters but V. K. Mittal, “A voice-controlled multi-functional Smart Home
remains connected to a range of 20 meter as it Automation System,” 2015 Annual IEEE India Conference (IN-
is connected to Blue-tooth. The voice recognize DICON),New Delhi,2015,pp. 1-6.
[7] M. Katore and M. R. Bachute, “Speech based human machine
accuracy achieved is 85.25% with four parame- interaction system for home automation,” 2015 IEEE Bombay
ters of access groups.The system also helps the Section Symposium (IBSS),Mumbai, 2015, pp. 1-6.
elderly and physically challenged people. The Blue- [8] A. R. Behera, J. Devi and D. S. Mishra, “A comparative study
and implementation of real time Home Automation System,” in
tooth Client was successfully checked on a level of 2015 International Conference on Energy Systems and Applica-
different smart-phones from various manufacturers, tions,Pune, 2015,pp. 28-33.
thus proving its portability and wide compatibil- [9] MIT App Inventor, “http://appinventor.mit.edu/,” July 2016.
ity. With this, People Counter makes the HAS [10] W. Yang, Y. Wu and G. Chen, “Application of Voice Recogni-
tion for Mobile E-Commerce Security,” in 2011 Third Pacific-Asia
more convenient to use. The prototype developed Conference on Circuits, Communications and System (PACCS)
can control the offices, homes and automobiles. ,Wuhan, 2011,pp. 1-4.
These enhancements have enhanced the detection [11] N. bt Aripin and M. B. Othman, “Voice control of home appli-
ances using Android,” in 2014 Electrical Power, Electronics, Com-
and made the system robust,error free and efficient. municatons, Control and Informatics Seminar (EECCIS),Malang,
MIT app inventor, an open free software has made 2014,pp. 142-146.
it possible to make an app which realized difficult [12] M. Bayin and M. Karakse, “Real time voice recognition based
smart home application,” in 2012 20th Signal Processing and
tasks quite easy. Also Arduino uno, made it easy to Communications Applications Conference (SIU),Mugla, 2012,pp.
use because of its enriched libraries. 1-4.
[13] M. Asadullah and A. Raza, “An overview of home automation
systems,” in 2016 2nd International Conference on Robotics and
VI. F UTURE S COPE Artificial Intelligence (ICRAI),Rawalpindi, 2016, pp. 27-31.
[14] R. Piyare and M. Tazil, “Bluetooth based home automation
This technology can be further developed by system using cell phone,” in 2011 IEEE 15th International Sympo-
integrating it with Internet while controlling it in sium on Consumer Electronics (ISCE),Singapore, 2011, pp. 192-
the remote area. This is beneficial in controlling the 195.
[15] G. C. Harikiran, K. Menasinkai and S. Shirol, “Smart security
load, appliances, gadgets and robots where human solution for women based on Internet Of Things(IOT),” in 2016
intervention is not possible. This will not only International Conference on Electrical, Electronics, and Opti-
improve security but will also assist in conservation mization Techniques (ICEEOT),Chennai, 2016, pp. 3551-3554.

578