VOICE ASSISTANT USING PYTHON

A DESIGN PROJECT - 1 REPORT

Submitted by:
DORESWAMY S B (2022BCSE07AED956)
MANASVI S (2022BCSE07AED957)
POORNIMA (2022BCSE07AED953)
NITYA (2022BCSE07AED958)
DARSHAN (2022BCSE07AED954)

In partial fulfilment for the award of the degree of

BACHELOR OF TECHNOLOGY

IN

COMPUTER SCIENCE AND ENGINEERING

Under the Supervision of

Dr. ASHWINI JAGANATH

Department of Computer Science and Engineering

ALLIANCE COLLEGE OF ENGINEERING AND DESIGN
Alliance University - Bangalore-562106
APRIL- 2024

1|Page
 Department of Computer Science and Engineering
Alliance College of Engineering and Design
Alliance University
Chikkahagade Cross, Chandapura-Anekal Main Road, Bangalore-562106

CERTIFICATE
This is to certify that the Design Project – 1 work entitled “VOICE ASSISTANT USING
“PYTHON” is the bonafide work done by Mr. DORESWAMY S B
(2022BCSE07AED956), Ms. MANASVI S (2022BCSE07AED957), Ms. POORNIMA
(2022BCSE07AED953), Ms. NITYA (2022BCSE07AED958), Mr. DARSHAN
(2022BCSE07AED954) submitted in partial fulfillment of the requirements for the Award
of the Degree Bachelor of Technology in Computer Science and Engineering during the
year 2022- 2026.

Dr.Ashwini Jagannath A. EZIL SAM LENI

[SUPERVISOR] [HOD]

External Examiners:
1. Name: Signature
2. Name: Signature

2|Page
 APPENDIX 3

TABLE OF CONTENTS

CHAPTER TITLE PAGE

NO. NO.
1 Introduction 6
2 LITERATURE REVIEW 9
2.1 Comparision table with various parameters 10-13
2.2 2.2.1 Observations from the Literature 14
Review

2.2.2 Limitations of the literature review 15

3 3.1 PROBLEM STATEMENT 16

3.2 PROPOSED RESEARCH METHODLOGY 17
3.3 HARDWARE REQUIREMENTS 17
3.4 SOFTWARE REQUIREMENTS 17
REFERENCES 18
APPENDIX: PLAGIARISM REPORT 19

3|Page
 ABSTRACT

In today’s generation, where time is the important, and the demand for technology to simplify
and streamline our daily tasks has never been greater. One such application that gained
significant traction is voice assistant using phyton (VPA).

These applications are powered by artificial intelligence, it has an aim to help in various fields
and allow individuals to focus on more needful matters.

VPA has an ability to complete a wide range of tasks based on user input, typically through
voice commands. From searching for information of movies to writing notes and sending
messages and providing locations VPA place major role in changing all of these (based on
our digital lives).

In response to this growing demand, the development of this application using programing
languages like phyton has brought a much more gain in the area of research and
implementation.

By identifying the power of speech recognition and natural language processing these VPA’s
provide a user-friendly experience and help users to interact with their device with no effort.

Here this VPA’s gives knowledge about the languages power like by using phyton we can be
easily able to do the voice recognition and bring it up in very easy manner. Coming to phyton
it does has functionalities such as information retrieval, task management, and device control.

Trought the combination of tragic algorithms and user-friendly interfaces, this voice assistant
plays a productive and efficient role in our daily lives.

Voice-activated virtual assistants have become popular due to their easy way of using. These
applications powered by software programs have very interesting advantages, they carry out
specific wake words and carry out a wide range of tasks from sending messages on WhatsApp
to drafting emails.

This generation is characterized by the presence of smartphones and desktops, our lives have
become fast-paced and demanding. Keeping the word activity aside, this type of application
is making our lives easier by simplifying typing etc.

Driven by advancements in artificial intelligence and natural language processing, these

assistants can handle all the situations very well, by learning a python a person can build an
4|Page
application like voice assistant which is capable of executing complex task in an easy way.

We just wanted to know more about the voice assistant so here, we just aimed on exploring
the design and development of intelligent voice personal assistants using phyton, highlighting
their potential to simplify task, improve productivity and enhance user experience in today’s
digital age.

History
The history of voice assistants using Phyton application is a basement to the evolution of
technology and the growing demand for human and computer interactions While phyton is
popular language for various applications, it’s flexibility and ease of use have made it
particularly well-suited for developing VPA.
The journey began with early experiments in speech recognition and production, where
phyton libraries such as Speech recognition were used to make it as the basic voice
interactions with computer. These early implementations made a path to build complex
voice assistants.
Where machine learning is getting advanced these days, Phyton frameworks like Tensor
Flow and PY Torch became instrumental in training and deploying voice recognition
models. These models get paired with Phyton-based speech synthesis libraires like GTTS
(Google text-to-speech), enabled voice assistants to understand respond to user commands
with increasing accuracy.
The rise of smart speakers and virtual assistants from tech giants like Google, Amazon, and
Apple. The demand for voice-enabled applications gained popularity. Phyton’s broad
information provided developers to create integrations and skills for these platforms,
expanding the capabilities of voice assistants to encompass a wide range of tasks, form home
autorotation to productivity tools
Today, Phyton continues to be the first application of voice assistant development, providing
innovative applications that transfer cutting-edge technologies like natural language
understanding, machine learning and cloud computing. As this field continues to evolve,
Phyton’s role in building the voice assistant will be difficult.
Looking ahead, Phyton is able to continue shaping the future of voice assistant technology.
With ongoing advantages on machine learning, NLP, and cloud computing, Phyton based
voice assistants are expected to become even more intelligent, responsive and integrated in
our daily lives.

5|Page
 CHAPTER 1
INTRODUCTION

A Voice Assistant application is a very helpful application which makes our life very easy by
converting complex things into simplified ways. So, the point comes how does it function?
So, the answer is by processing spoken commands from users it does its actions. When a user
speaks into the device’s microphone, the voice assistant employs search recognition
technology to convert the spoken words into text.
Once the user’s intent is discerned, the voice assistant determines the appropriate action to
fulfill the request, whether it’s retrying information from the web, controlling smart home
devices, or performing other tasks

VPA has an ability to complete a wide range of tasks based on user input, typically through
voice commands. From searching for information of movies to writing notes and sending
messages and providing locations VPA place major role in changing all of these
A Voice assistant application operates by understanding and responding to spoken commands
by user. When user speaks into the microphone of their device, the voice assistant uses the
advanced speech recognition technology to accurately transcribe the spoken words to text
And here we just wanted slight change and thought of bringing a voice detector like which
recognizes or voice and tell our mood swing and bring up the suitable songs or some activity
to make up the mood better, if we are in mood of happy there comes the motivation which
keeps us moving forward like this.

So here we just thought of this project, which will be helpful for the people who are suffering
from their mental illness, where they can’t tell their problems to the human beings, can have
it and I'm not sure about making the person complete fine, but yeah suffering alone will be
degraded because this app will be giving the enough support for their stability for upcoming
days.

We just looked at the applications which were developed by developers, so we can bring our
own idea as I informed above, we thought of making a brief app like collecting data of 50
mood swings and bringing the correct solution for this. We discussed it with our mentor too
and it is in the process.

In this generation this application brings happiness to everyone, including the child, teen
youths, everyone. So, Voice assistant app makes up the major role and even the feelings
detector would help we guess.

6|Page
 CHAPTER 2

REVIEW OF LITREATURE

Author Name,
Journal, Research Performance Advant Disadvant
S.No
Publication Methodology Metrics ages ages
Year

1. P Srinivas T The purpose of this Python is an Ease of Limited Control:

Sai Teja Pub paper is to demonstrate interpreted Development: Python's high-
year:2020 the implementation of language, Python's level nature may
International a voice user interface which can simplicity limit fine-grained
Journal of (VUI) as a personal result in slower makes it easy control over
Engineering voice assistant that can performance to develop system-level
Applied perform several compared to voice functionalities,
Sciences and assignments or tasks compiled assistants, which could be
Technology 4 for the user such as languages like even for crucial for some
(11), 105-108, setting up the alarm, C++ or Java, beginners. advanced voice
2020 reminders for the day, especially for assistant feature
knowing the weather computationall
forecast, reading news y intensive
feed, to play a song tasks.
from the playlist, ask
about the details of a
movie, finding out the
meaning of an
unknown word, to read
a article from
Wikipedia,

V Rajesh, Describe a virtual Matrices should cross-platform Platform

2. personal assistant for have included Compatibility: Dependencies:
K Saikumar smart phones and other Latency which meaning voice While Python
devices that acts as a measures the assistants built itself is cross-
Pub year:2021 helping hand to delay between with Python platform, certain
3rd improve various when a user can run on libraries or
International personal speaks a various modules used in
Conference on responsibilities, command and operating voice assistant
Advances in allowing people to when the voice systems development may
Computing, focus on more essential assistant without much have platform-
Communicatio matters. It completes responds. specific
7|Page
 n Control and the assignment based Lower latency modification. dependencies,
Networking on a variety of leads to a more leading to
(ICAC3N), different questions responsive user compatibility
1650-1654, provided by the voice experience. It issues across
2021 assistant. can be affected different
by factors such operating
as speech systems.
recognition
speed, natural
language
processing
time, and
system
responsiveness

S Akash, A voice assistant is Accuracy Integration Callability

3. Neeraj Pub also known as voice- matrices refers with Web Challenges: As
year:2022 6th based Artificial to the voice Technologies: voice assistants
International Intelligence. It is one assistant's Python's web grow in
Conference on of the hot topics in the ability to frameworks complexity and
Intelligent current world, that are accurately like Django user base, scaling
Computing programs that listens understand and and Flask Python-based
and Control to human’s verbal interpret user facilitate the applications may
Systems command and commands. integration of become
(ICICCS), responds to them This includes voice challenging due
374-379, 2022 which makes it a both speech assistants with to limitations in
human-computer recognition web services, concurrency,
device interaction. accuracy enabling performance, and
(transcribing features like resource
spoken words voice- utilization.
correctly) controlled web
Higher browsing or
accuracy leads accessing
to more online
effective information.
interactions and
fewer errors.

8|Page
 4
Divisha Pandey Started by This includes It provides rich Recourses
Researching both speech Ecosystem; it has consumption,
Afra Ali existing voice recognition a vast ecosystem Phytons dynamic
assistant accuracy library and typing and
Pub year:2022 mythologies. We (transcribing framework for automatic memory
studied this paper spoken words natural language management can
deeply and brought correctly) processing. lead to higher
out these ideas. Higher resource of
accuracy leads consumption .
to more.

5. Anjali Fahal, The main goal of We use python Python offers a Python, while easy
Trupti Kanade Artificial as a wide range of use and flexible,
intelligence (AI) is programming libraries and might not be the
Pub year:2021 the realization of language Frameworks for fastest language for
International natural dialogue because it have natural language CPU-intensive tasks.
Research Journal between humans a major processing(NLP) However, for most
of Modernization and machines. libraries which and other Ai personal assistant
in Engineering There are many IT is use to related tasks. Applications.
companies have execute
used the dialogue commands. By
systems using python
technology to Installer
establish various Packages our
kinds of Virtual personal virtual
Personal

9|Page
 6. Neha Prasad, In a world with In the following The Integration Reliance on
Pranaya Kodava ever increasing paper, we of chatbot and Internet Connection:
Pub year:2020 needs for comfort, propose a Both chatbots and
ITM Web of human race is Home Voice assistant voice assistants
Conferences relying more and Automation technology require a stable
32,01002,2020 more on system that allows for hands- internet connection
technological uses a free and natural to function properly.
advancements to wholesome language control Any disruption in
find solutions to blending of of smart home the internet service
their problems. some devices. could affect the
technologies usability of home
like Internet of automation. Privacy
Things. Concerns: Voice
assistants often raise
privacy concerns as
they continuously
listen for trigger
words. Users may be
hesitant to have
devices listening to
their conversations
at all times.

7. Ahmed J This paper Python is Enhanced Accuracy and

Abougarair, approaches the use chosen as a Academic Recognition
Mohamed K of a Virtual development Support: A voice Challenges:
Aburakhis, M Assistant using language due to assistant that Academic
Zaroug neural networks its capabilities recognizes terminology can be
International for recognition of and academic words highly specialized
Robotics & commonly used compatibility can provide real- and context-
Automation words. The main with various time definitions, dependent. Voice
Journal purpose is to APIs and explanations, and assistants may
facilitate the users’ libraries, which context for struggle to
daily lives by are deemed complex terms accurately
sensing the voice necessary for commonly used recognize and
and interpreting it the project. The in academic interpret less
into action. virtual assistant settings. This can common or context-
will be required greatly enhance specific academic
to communicate learning and terms.
with IoT comprehension.
devices.

10 | P a g e
2.2.1 OBSERVATIONS FROM LITERATURE REVIEW
• Brief Description of observations made from the papers.

Accuracy of Speech Recognition: Assessing how accurately the voice assistant

recognizes spoken commands or queries. This can involve noticing instances where the
assistant correctly understands and responds to the user's voice inputs, as well as cases
where it misinterprets or fails to understand them.

Response Time: Evaluating the speed at which the voice assistant processes voice inputs
and provides responses. Faster response times generally lead to a more seamless user
experience, while delays can be frustrating.

Natural Language Understanding: Observing the assistant's ability to comprehend and

respond appropriately to natural language queries and commands. This involves
assessing how well it understands context, synonyms, and variations in language.

Error Handling: Noting how the voice assistant handles errors or misunderstandings.
This includes observing whether it provides helpful prompts or clarifications when it
doesn't understand a command, and how it handles situations where it cannot fulfill a
request.

Integration with Other Services: If the voice assistant interacts with external services or
APIs, observing how well it integrates with those services and the accuracy of the
information it retrieves or actions it performs.

User Interface: If the voice assistant has a graphical user interface (GUI) or other means
of interaction beyond voice, assessing the usability and intuitiveness of these interfaces.

Feedback Mechanisms: Examining how the voice assistant provides feedback to the
user, such as through spoken responses, visual cues, or notifications.

Customization and Extensibility: If the voice assistant allows for customization or the
addition of new features, observing how easy it is to extend its functionality and tailor
it to specific needs or preferences.

11 | P a g e
2.2.2 LIMITATIONS OF LITERATURE REVIEW

Brief Description of the limitations observed from the papers.

1.The limitations we observed from the paper include dialectical variability which
affects the performance. the higher accuracy has been holding by male spoken digits
compared to female spoken digits in gender – based experiments. Including that, this
paper study mentions the major requirement of algorithm of a specific size and
recording the circumstances which is recorded in speech, which is one of the limitations
in strategy. The Paper we referred highlights the importance of integrating machine
learning and deep learning to grace the performance of voice assistants and control
various tasks which is efficient.

2.Though only limitations do not bring up the solution, as we should take up any of the
solution, we call it has possibilities of overcoming on the limitations. To overcome the
limitations which are given in the paper, firstly we should focus on integrating machine
learning and deep learning techniques to build the performance of voice assistants and
control the tasks efficiently. Adding to it, Addressing the di electrical

3.Variability by considering the difference in accuracy between female and male spoken
digits in gender-based experiments could have brought up to overcome from limitations
Possibilities of overcoming on the limitations in this proposed work

12 | P a g e
 CHAPTER 3

PROBLEM DEFINITION & PROPOSED RESEARCH METHODOLOGY

3.1 PROBLEM STATEMENT

Statement of the problem

In today's age of technology, there is a great value for quick and brilliant voice
assistants that can make the task easy and give a productive solution. However,
improving a voice assistant in Python has few challenges, which includes natural
language understanding, speech recognition, context awareness, and closest possible
co-ordination with various applications. The aim of this project is to design and
implement a advanced voice assistant using Python that can understand and respond
to user commands and perform tasks such as scheduling appointments, collecting
information from the web, controlling smart home devices, and more, all while
assuring user privacy and data security

In the present internet era, in person and business are much reliant on technology to
do daily tasks and improve productivity. As part of this digital evolution, there is
increasing value for voice assistants that can co-ordinately integrate into various
applications. However, developing an improved voice assistant in python presents
multifaced challenge is a challenging task for the developers.

Developing a voice assistant using python must also include some tasks that can be
easily done by it for example: setting a reminder, alarm, playing music, and many
more. It must be able to understand the languages that are used by the individual.
And the aim is to make is easy for the uneducated people to use the technology and
help to solve their queries.

13 | P a g e
3.2 PROPOSED RESEARCH METHODOLOGY

• Detailed explanation of the proposed work:

The researched methodologies for improving a voice assistant using python can be
approached through a merge of theoretical thinking and practical experiment and
good development. Some of the proposed methodologies are:
1. Literature Review
2. Problem Definition
3. Data Collection
4. Algorithm Selection
5. Documentation and Reporting

The proposed methodologies for the given problem statement involve

3.3 HARDWARE REQUIREMENTS

Hardware Specification of the system: To develop a voice assistant we need a
hardware component that supports both speech recognition and audio processing.
Some of the hardware components are:

1. Computers/Laptops
2. Memory
3. Microphones
4. Speakers
5. Sound Card
6. Internet Connection.

3.4 SOFTWARE REQUIREMENTS

Tools and Technologies used for solving this proposed problem / the solution.
To enhance a voice-assistant we need a software component that supports both
speech recognition and audio processing. Some of the software components are:

1. Speech Recognition Library

2. Natural Language Processing (NLP) Library
3. Text-to-Speech (TTS) Library
4. Assistant Logic
5. Development Tools.

14 | P a g e
REFERENCES (MIN. OF 10 PAPERS)
1.M. Bapat, H. Gune, and P. Bhattacharyya, “A paradigm-based finite state morphological
analyzer for Marathi,” in Proceedings of the 1st Workshop on South and Southeast Asian
Natural Language Processing (WSSANLP), pp. 26–34, 2010.

2. B. S. Atal and L. R. Rabiner, “A pattern recognition approach to voiced unvoiced-silence

classification with applications to speech recognition,” Acoustics, Speech and Signal
Processing, IEEE Transactions on, vol. 24, no. 3, pp. 201–212, 1976.

3. V. Radha and C. Vimala, “A review on speech recognition challenges and approaches,”

doaj. org, vol. 2, no. 1, pp. 1–7, 2012.

4.T. Schultz and A. Waibel, “Language- independent and language adaptive

acousticmodeling for speech recognition”, SpeechCommunication, vol. 35, no. 1, pp. 31–51,
2001.

5. J. B. Allen, “From lord rayleigh to shannon: Howdo humans decode speech,” in

International Conference on Acoustics, Speech and Signal Processing, 2002.

6. M. Bapat, H. Gune, and P. Bhattacharyya, “A paradigm-based finite state morphological

analyzer for marathi,” in Proceedings of the 1st Workshop on South and Southeast Asian
Natural Language Processing (WSSANLP), pp. 26–34, 2010.

7.G. Muhammad, Y. Alotaibi, M. N. Huda, et al., pronunciation variation for asr: A survey
of the “Automatic speech recognition for digits, literature, Speech Communication, vol. 29,
no. in Computers and Information Technology, 2009.2, pp. 225–246, 1999.

8. S. R. Eddy, “Hidden Markov models,” Current opinion in structural biology, vol. 6, no. 3,
pp. 361–365, 1996.

9. Excellent style manual for science writers is “Speech recognition with flat direct models,”
IEEE Journal of Selected Topics in Signal Processing, 2010.

10. Srivastava S., Prakash S. (2020) Security Enhancement of IoT Based Smart Home Using
Hybrid Technique. In: Bhattacharjee A., Borgohain S., Soni B., Verma G., Gao XZ. (eds)
Machine Learning, Image Processing, Network Security and Data Sciences. MIND 2020.
Communications in Computer and Information Science, vol 1241. Springer, Singapore.
https://doi.org/10.1007/978-981-15-6318-8_44

15 | P a g e
APPENDIX: PLAGIARISM REPORT

16 | P a g e