M. Brindha , K. Deepika , N. Aruna, K.S. Harini, Mrs. S. Prince
SahayaBrighty
Student, Department of Computer Science and Engineering, Sri
Ramakrishna Engineering College, Coimbatore, Tamil Nadu, India.
Assistant Professor, Department of Computer Science and Engineering,
Sri Ramakrishna Engineering College, Coimbatore, Tamil Nadu, India.
ABSTRACT
The main
aim of artificial intelligence is to allowassignation with electronic devices and
computer in a simple and practical way. Personal assistant which can do any
task just by listening a phrase is trending now a days. A virtual assistant is
able to understand and give reply to human commands using synthesized voices.
Many organization have created Virtual Personal Assistants (VPAs) based on
their requirement and domains, such as Cortina
which is made by microst,Siri which is made by apple, Alexa by amazon, and
Google Assistant by Google, using dialogue systems technology. User can give
commands to perform task such as play the music , as well as can do the basic tasks like messages, to-do lists, and so
onHowever, the generally works online but our research work can do the work in
both mode online as well as offline. So here we introduce Vizhi (An AI
Assistant) with Voice Commands Intelligence, who receives the user command in
the form of voice or text thenvisit process it and give the output such as the
task which is said to be completed.
Keywords— Natural Language Understanding, Artificial Intelligence, Application
Program Interface, Virtual Personal Assistant, User Interface.
I. INTRODUCTION
People in today's world are worried
out as a result of their fast-paced, stereotyped lives. Peoples is not having a
much time for themselves because of daily routine work for that they travel a lot, and buy a lot. The purpose of
this epoch is to make everything more efficient, simple, and quick. People may
have simply gotten lazy through time, but why do anything when we can have
something else do it for us? As a result of this, as well as technological
advancements, several virtual assistant implementations are now accessible.
A virtual assistant, also
known as a digital assistant or an AI helper, is a piece of software or an
application program that can converse with the user in common language and
comprehend what he says. A virtual assistant, also known as a digital assistant
or an AI helper, is a piece of software or an application program that can
converse with the user in common language and comprehend what he says.
Speech-to-text and
text-to-speech transformations are vital for the process, as are several AI
systems based on machine learning, deep learning, and neural networks, which
act as a brain for these assistants.
·
Vichy
isn't simply another virtual assistant. You may journal, discuss your day,
remind yourself of tasks, and even lighten your day with encouraging remarks
and corny jokes.
·
Interactive
lighthearted conversations may be calming and beneficial for a variety of conditions.
It's like chatting with your best buddy through a texting app.
In most cases, a user
must manually manage many sets of programs in order to execute a single
operation. A system that can effortlessly handle chores is required. We already
have a number of virtual assistants on staff. However, we barely ever utilize
it. A large proportion of people struggle with speech recognition. These
systems are capable of understanding they can understand English sentences, but
not our accent. Our way of speaking is different from there. They are easy to
use on smartphones compared to desktop. A virtual assistant is able to
understand English in different accents and work without internet connection is
required to meet the demands of users. In order for them to function well, they
must be given a tremendous volume of data.
·
It
stands out from the crowd of popular helper applications. Others may take a few
seconds to answer, but Valhi responds instantly. Furthermore, unlike other
virtual assistants, it may run in offline mode, conserving your data.
·
Vichy
works in tandem with your other applications and services.
·
You
activate it with a simple command, similar to Google Assistant: "Hey Valhi"
or "Hey," or any phrase you choose to use as a key word.
DISTINCTIVE IN VIZHI:
§ Absolute integration.
§ Accents varying.
§ Quick response.
§ When you are unwell or sad, it
provides you with some recovery tips and professionals to call.
§ Permit the blind propel to engage
(only limited aspects).
§ Simple User Interface
§ Offline features for journals.
The Android application,
which implements and displays all of the functionality, is built in Eclipse
using Android development tools.To capture incoming requests, the application
uses speech recognition. Creating the main activity and each of the functions,
as well as applying the logic to put the entire program together. Furthermore,
the command may be examined with the database storage by querying the web
service on firebase; relevant replies will be directed to certain functions in
the application.
II. LITERATUREREVIEW
UNDERSTANDING OF VPA’S: VPA working is described in
the figure below.
Fig.1 VPA
Voice Recognition:
·
The
android phone will record the voice input in the first step.[1]
·
Whenever
the people will say anything in their voice which is natural, the assistant will
be able to transform the analog signal into the digital signal, and then
combine all after arranging them in proper keywords and then giving the result
after following the user commands.
Understand what the statement
implies:
·
People enjoy playing with new things, so it's
no surprise that they'll say anything that comes to mind.
·
As a result, AI assistants are able to
analyses the context of a statement using Automatic Speech Recognition (ASR),
Natural Language Processing (NLP), Desired business logic through hooks, and
Text To Speech (TTS) and process them appropriately. [1]
·
•
Android applications will identify the voice utilizing the Android API and Java
API. A recorded text will be made and transferred to the cloud server or
Android applications, depending on the instruction.
Action based on what was commanded:
·
Using
the Java API, references, and a preconfigured database, the cloud server (Fire
base) will decode the incoming text [9], then determine which of
the following steps should be carried out.The cloud server will convert a command into a URL and
send it to the appropriate server (Google server, Wikipedia server).
·
The
server that receives the request will create the answer in XML or JSON format
using the Wikipedia API or Google API.
·
Convert
to a particular answer that will be directed to the Android app.
The Android programmer will create
the audio output, which will be communicated to the user via the mobile
speaker. The development of voice recognition technology has had a long and
tortuous path. Nonetheless, without the early pioneers who led the way, today's
voice systems such as Google Voice, Amazon Alexa, Microsoft Cortana, and
Apple's Sire [3] would not be where they are today.
These speech systems have
increased their capacity to 'hear' and comprehend a greater range of words,
languages, and accents as a result of the integration of new technologies such
as cloud-based processing and continuing data gathering programmers.
This project focuses on
Android voice control (recognition, generation and analysis of corresponding
commands, intelligent responses automatically), Google products and relevant
APIs (Google map, Google weather, Google search, and so on), Wikipedia API, and
mobile device references ranging from Speech-To-Text, Text-To-Speech
technology, and camera; Vichy will be able to relate words to one another in
terms of adjective, verb, noun and with full sentence. All of the project's
functions and services have been discussed.
Fig.2 Action Diagram of VPA
III. SYSTEM DESIGN AND
IMPLEMENTATIONBLOCK DIAGRAM:
·
To
take orders from the user, the model uses human input through microphone.
·
After
that, these directives go via the potential of computer and program to
understand the spoken words and translate them machine understandable format which
is known as speech recognition.
·
Natural
Language Processing is used on these inputs; it is a field that combines
computer science and artificial intelligence. We are using Natural Language
Processing (NLP) to concern with interactions between computers and human
natural languages.
·
If
it's a question, Vizhi determines if the statement is an inquiry or an action.
If it's an action, the voice assistant will carry it out, and the user will be
notified via synthetic voice. If you have a question, Vizhi searches the
dialogue box or knowledge base and responds to the user via a synthesis voice.
·
Our
voice assistant analyzes all of the user's utterances using the Google
Text-to-Speech API, and then provides replies based on particular situations
that qualify as a directive to the user.
IV. DATA FLOW DIAGRAM:
Fig. 3 DFD of VPA
·
The
user gives the instruction through the voice which the system fetch.
·
If
it is determined that the user is authorized, action is taken in accordance
with the user's instruction.
·
The
Google Speech API will transform this voice data to text, and the voice
assistant will then carry out the instruction given by the user by comparing it
to the dialogue box and knowledge base.
V. USE CASE
DIAGRAM:
Fig. 4 Use Case of VPA
VI. APP UTILIZATION:
Entertainment🎶
·
Play
music
·
Stop/start
music
·
Crack
a joke ,
Creativity ✔️
·
Check
my schedule
·
Send
messages to any one
·
weather
forecast
·
send
messages throughWhat Sapp
·
Discover(anything)
Checklist⏰
·
alert
me to [reminder] at [time] on [day]
·
Recall
[something] So, what do you recall?
·
Forget
whatever I told you to remember
·
journal
reading/writing clean up my journal entry set a timer
Navigation🌍
·
To
get to [location], call an Uber.
·
I'm
hungry and need instructions to [location].
Device functions📲
·
Snap
a photo
·
On/off
Bluetooth
·
Wi-Fi
on and off
·
Call
by name
Miscellaneous📈🎲
·
Tell
me some facts
·
Toss
a coin
·
Repeat
the process
·
express
[what you want to be expressed]
·
alter
your accent
Fig. 5 Accuracy Graph of VPA
VII. RESULTS
VIII. ADVANTAGES AND DISADVANTAGES
ADVANTAGES
ü Work will be done both offline and
online..
ü It is simple to use and accessible to
people of all ages..
ü Changeable voices.
ü
Quick
response
ü
a
new batch of cheesy entertaining facts and jokes.
ü
Open
your current apps.
ü Compared to other VPAs, it uses less
data.
DISADVANTAGES
v Only the English language is
compatible.
v Voice is more relevant
than graphics interface.
IX. CONCLUSION
Vizhi will play an important role and
be with you like buddy and stress reliever. Users can understand that, no
matter how hard developers strive to add more predefined commands, more reply
to them, and evaluate and respond to instructions more intelligently, the
software will never be totally comprehensive and encompass all of the
conceivable scenarios that users will encounter. However, if there are more
understandable instructions, a more humanized structure, and a more intelligent
response, the software will undoubtedly improve and become more user-friendly.
The interface design now fits the fundamental criteria of presenting everything
for this software, and users may interact with it through it, but it will be optimized
and more suitably developed in the future.