International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056

Volume: 11 Issue: 04 | Apr 2024 www.irjet.net p-ISSN: 2395-0072

VOICE ASSISTANT: DESKTOP-BASED APPLICATION

Jay Mhatre1, Prasanna Tayare2, Sanjeev Kumar3, Pratham Temkar4, Dr. Mahendra Pawar5

1. Student, Dept. of COMP Engineering, VPPCOE & VA, Maharashtra, India

2. Student, Dept. of COMP Engineering, VPPCOE & VA, Maharashtra, India
3. Student, Dept. of COMP Engineering, VPPCOE & VA, Maharashtra, India
4. Student, Dept. of COMP Engineering, VPPCOE & VA, Maharashtra, India
5. Associate Professor, Dept. of COMP Engineering, VPPCOE & VA, Maharashtra, India
---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - With voice assistants emerging as essential flexibility and user-friendliness. We will leverage Python's
elements of advanced computer systems, they provide a powerful libraries and frameworks for natural language
convenient and user-friendly way to interact with the processing and speech recognition to achieve this goal.
technology. This research project concentrates on creating Our primary aim is to create an automated personal
and using an Electron JS and Python-based voice assistant assistant that enhances user productivity and convenience
that is desktop-oriented. The work investigates technical while demonstrating the feasibility of this technology. Our
architecture, development process, and user experience prototype will showcase basic functionalities, laying the
aspects associated with designing a desktop-based interface groundwork for future advancements and optimizations.
driven by voice. Using Electron JS allows for compatibility
across platforms as well as incorporating web technologies 2. BACKGROUND
while Python provides core features like speech recognition
and natural language processing. This project reveals some 2.1. History of Voice Assistant
aspects of creating a voice assistant for desktop use such as
speech accuracy, UI design, system integration, etc. From simple speech recognition software to advanced AI-
Feedback from users and testing results help shed light on powered virtual assistants, voice assistants have come a
the usability and efficiency of the voice assistant application. long way. The voice assistant journey, however; had
Finally, this paper discusses possible future directions for started in the mid-20th century with crude attempts at
improvements to make these types of tools more widely speech recognition. By the 1950s Bell Labs had introduced
applicable in other areas beyond desktop assistance. The “Audrey” then by the 1960s IBM came up with “Shoebox”.
contribution made by this study is focused on how to These served as a basis for further development.
advance the field of voice assistant software toward being Nonetheless, these earliest systems were limited by
integrated into desktop environments. technology and failed to gain widespread use due to their
inadequacies and limited accuracy. This took place at the
Key Words: Voice Assistant, Recognition, NLP, Desktop turn of the century when more advanced voice assistants
were being developed. As early as 2001, Microsoft’s
1. INTRODUCTION “Cortana” was launched, which led to Apple’s release of
In recent years, voice assistants have revolutionized how Siri on iPhone 4S in 2011. Introduced via iPhone 4S, Siri
we interact with technology, providing a convenient and ensured that everyone bought into the virtual assistant
natural way for people to engage with their devices. concept through natural language command hence
Today, these assistants have become an integral part of starting an era of voice-controlled interface. These are
our daily lives, assisting with everything from reminders among some of the voice assistant platforms that have
to controlling smart home devices. While voice assistants dominated the market such as Amazon’s Alexa in 2014
have traditionally been linked to portable and home-based and Google Assistant released in 2016. These platforms
electronics, there is a growing demand to expand their used artificial intelligence (AI), natural language
functionality to desktop computer settings. This research processing (NLP), and cloud computing so that they
project aims to create and deploy a desktop voice assistant became increasingly personalized and responsive. Today
application using Python and Electron JS. The proposed there are voice assistants implemented across various
voice assistant seeks to enhance accessibility and devices or services thereby transforming how users
productivity by enabling hands-free interaction with interact with technology every day.
desktop computers. Unlike traditional voice assistants that
are primarily made for mobile or smart speaker platforms, 2.2 Future Applications
this project will use web technologies like HTML, CSS, and
JavaScript to develop cross-platform desktop applications Some other future applications of voice assistants are
using Electron JS. Our objective is to develop a voice personalized healthcare support, advanced residential
assistant that is optimized for desktop use, with a focus on automation with predictive capabilities, immersive virtual

© 2024, IRJET | Impact Factor value: 8.226 | ISO 9001:2008 Certified Journal | Page 2057
 International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 11 Issue: 04 | Apr 2024 www.irjet.net p-ISSN: 2395-0072

reality experiences with natural language interaction, and

enhanced accessibility features for disabled people. By this
means, the voice assistant will become more embedded in
daily life, offering convenience and help across a range of
contexts.

2.3 Aim of this Study

Our research aims to pave the way for more efficient and
natural human-computer interactions by integrating voice
assistant technology into desktop computing
environments.

3. PROPOSED DESIGN
The project will provide a reasonable grasp of an
intelligent assistant that can comprehend user commands.
Our voice assistant recognizes and complies with
commands given by the user through vocal media and
responds as required. We develop a prototype of an Figure 1: Use-Case Diagram
automated personal assistant capable of understanding
In this diagram, labeled as Figure 1, we can see the Use-
user commands and performing tasks across various
Case illustration showcasing the interaction between the
domains, including scheduling, information retrieval, and
user and the voice assistant. The user initiates the
task automation. Using web technologies, the proposed
conversation by speaking to the assistant, who then
desktop voice assistant application will be built cross-
proceeds to identify their speech, interpret their
platform for desktop apps using the Electron JS
intentions, carry out relevant tasks, and generate a fitting
framework. Using HTML, CSS, and JavaScript, Electron
response. The assistant has to furnish the requested
offers a single development environment that makes it
information or perform the necessary action.
easy to integrate with current web-based technologies
while developing desktop apps. Python will be used to
implement the application's backend features, which 4. METHODOLOGY
include speech recognition and natural language
processing. Python provides strong frameworks and tools When a user accesses a Voice Assistant application, they
to perform these tasks such as pyttsx3, are presented with a desktop application that allows them
SpeechRecognition, Pyaudio, etc. to interact with the platform. The backend functionalities
of the assistant were built using Python, while web
Our study aims to achieve several important goals, technologies such as HTML, CSS, and JavaScript were used
including: to create the frontend interface through Electron JS. To
enable communication between the frontend and backend
 To investigate the principles of design and components, FastAPI was implemented. Various modules
technical architecture of desktop voice assistants. were developed and integrated using specific technologies
 To look at the difficulties and development and frameworks to support the Voice Assistant's basic
process involved in creating a voice-activated operations, AI, voice recognition, face recognition, emotion
desktop interface. identification, and system application control. To enable
 To investigate the factors related to user system voice within the program, we utilized Sapi5 and
experience and usability when utilizing a desktop Pyttsx3. The latter is a Python text-to-speech conversion
voice assistant. module that operates offline. Additionally, we
 To test and assess the voice assistant application incorporated the Speech Recognition Library, which
to obtain user input and insights. provides numerous built-in functions. This library allows
 To determine future paths and possible users to send commands to the assistant and receive voice
improvements for desktop voice assistant responses in return, complete with Text-to-speech
technology. capabilities. Upon recording, the assistant's underlying
algorithms convert the user's voice command to text.
When a user gives a command such as "Open Notepad" or
"Play a Song on YouTube", the voice assistant responds by
executing the command. The assistant waits for a pause to
indicate that the user has finished speaking, then searches

© 2024, IRJET | Impact Factor value: 8.226 | ISO 9001:2008 Certified Journal | Page 2058
 International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 11 Issue: 04 | Apr 2024 www.irjet.net p-ISSN: 2395-0072

its database for the user's command and executes it and mapped to an action. The assistant executes the action
accordingly. and generates a response, which is spoken to the user.
 The user's request is split into separate
commands to aid understanding by the voice 5. IMPLEMENTATION & RESULTS
assistant.
 Once the command list is accessed, it searches and "Sifra," the voice assistant program for desktop use, is
compares our requests with others. designed with an intuitive user interface. When the
 The list of commands is then sent back to the application is launched, users will be presented with a
voice assistant. login page that utilizes facial recognition for
 After receiving the commands, the voice assistant authentication. This system is crucial in ensuring user
will know what action to take next. identification and secure access to the Assistant interface.
 The voice assistant will ask for clarification if a Authenticating user credentials, limits access to only
request is unclear to ensure it understands the authorized users who can interact with the Assistant and
user's desired outcome. execute commands. Ultimately, this feature safeguards
 If the voice assistant can understand the user's user privacy and provides a trusted and secure interface.
request, it will perform the task.
In the final phase of distribution, the Voice Assistant was
deployed to desktop environments while adhering to
operating system standards and ensuring compatibility
with various platforms.

Figure 3: Flow Chart

Figure 3 shows a flowchart where a user activates the

voice assistant using a "wake-up" keyword. When the
voice assistant is activated, it will greet the user and wait
for the user to provide an input command. If the user
provides an input command through voice, the assistant
will capture it and search for any related keywords. If the
assistant finds a keyword related to the input command, it
will perform the task and return the output to the user
both in voice and in text form on the terminal window. If
the assistant cannot find any relevant keywords, it will
continue waiting for the user to input a valid command.
Each of these functionalities plays an important role in the
overall functioning of the system.
Figure 2: Sequence Diagram
The following tasks are performed by the Voice Assistant:
Figure 2 shows the sequence diagram that displays the  Launching applications, setting reminders,
interaction between the user and the voice assistant. The creating calendar events, sending messages, etc.
assistant wakes up on the wake word and listens to the  Searches anything from Google and tells the
user's request. The request is converted to text, analyzed, required content. If asked 'Google search' the

© 2024, IRJET | Impact Factor value: 8.226 | ISO 9001:2008 Certified Journal | Page 2059
 International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 11 Issue: 04 | Apr 2024 www.irjet.net p-ISSN: 2395-0072

assistant searches the content asked from Google Assigning Task To the Assistant:
and opens the required content in the browser.
 Tells the accurate weather of the location the user
asks for. When asked for 'current weather in' the
assistant tells the exact weather of the desired
location of both maximum and minimum in
degrees Celsius.
 Takes a screenshot of the display. When asked for
"capture", "capture my screen", "my screen",
"screenshot", or "take screenshot", the assistant
captures the display the user is using and stores it
in the path specified. Figure 5: Asking the assistant to open Notepad
 It sends a mail to the username specified by the
user. When told “send mail”, the assistant asks to
whom the mail has to be sent and it will send a
mail according to that.
 Can shut down or restart the system by just user
command etc.

Login through face recognition and activation of the

assistant:

Figure 6: Asking the assistant to search on Google

Figure 7: Asking the assistant to play music on

YouTube

6. CONCLUSION

In summary, making the Voice Assistant app for desktops

was a big effort to make computer use better with new
Figure 4: Opening Desktop Application tech. Using Electron JS and Python, the project made a
voice assistant that can do voice, face, and emotion
recognition, control system apps, do basic tasks, and
support artificial intelligence. By following a plan that
involved finding needs, doing tech research, designing,
making, testing, and writing stuff down, the project did
what it set out to do by giving users a smooth and easy
time. In the future, the Voice Assistant application has
vast potential for further advancements and
applications across multiple domains. It could

© 2024, IRJET | Impact Factor value: 8.226 | ISO 9001:2008 Certified Journal | Page 2060
 International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 11 Issue: 04 | Apr 2024 www.irjet.net p-ISSN: 2395-0072

understand language better, do more things, and work [8] A N Bhoomika Chowdary, Lavanya M, Darshan S,
with new tech. Overall, the Voice Helper project is a big Deepthi Andani, & Mrs. Jayasudha BSK “A Literature
step forward in how people and computers work Survey on Voice Assistant” IJRASET 2022.
together, making it easier and more personal to use
computers. [9] Faruk Lawal Ibrahim Dutsinma, Debajyoti Pal, Suree
Funikul, & Jonathan H. Chan “A Systematic Review of
Voice Assistant Usability: An ISO 9241-11 Approach”
ACKNOWLEDGEMENT
2022.
We would like to sincerely thank Dr. Mahendra Pawar, our
project guide, and the rest of the computer office's [10] S. Springenberg, M. Moberg, & O. Viikki, “Intelligent
teaching team for their significant suggestions, Personal Assistants” Speech Technology Seminar,
encouragement, and support during this research. Their Institute of Computer Science Hamburg, 2016.
knowledge and perceptions have been crucial in
determining the course and extent of this undertaking. We [11] V. Kukade, Ruchita G. Fengse, Kiran D. Rodge, Siddhi P.
also acknowledge the Department Head, Dr. Rais Mulla, for Ransing, & Vina M. Lomte, “Virtual Personal Assistant
his unwavering encouragement and assistance. His for the Blind”, International Journal of Computer
leadership has served as an inspiration to us. We also like Science and Technology, Volume 9, Issue 4, October-
to thank our peers and colleagues for their contributions, December 2018.
as they have offered insightful criticism and
recommendations. Their advice has substantially
improved our job. [12] Ritik Porwal, Ujjawal Tomar, Vishakha Dubey, Asst.
Prof. Akshita Mishra, Asst. Prof. Gourav Mandloi
REFERENCES “Voice Assistant” ISSN 2021.

[1] Subhash S, Prajwal N Srivatsa, Siddesh S, Ullas A, &

Santosh B “Artificial Intelligence-based Voic-Assistant”
IEEE 2020.

[2] Ms. Preethi G, Mr. Thiruppugal, Mr. Abhishek, & Mr.

Vishwas D A “Voice Assistant using Artificial
Intelligence ” IJERT 2022.

[3] G Rushivardhan, & Mrs. K Santoshi “Virtual Assistant

using NLP Techniques” IJCRT 2022.

[4] Veton Kepuska, & Gamal Bohouta “Next-Generation of

Virtual Personal Assistants (Microsoft Cortana, Apple
Siri, Amazon Alexa, and Google Home) ” IEEE 2018.

[5] G. Bohouta and V. Z Kepuska “Improving Wake-up-

Word and General Speech Recognition Systems”. IEEE
2017 15th Intl Conf on Dependable, Autonomic and
Secure Computing, 15th Intl Conf on Pervasive
Intelligence and Computing, 3rd Intl Conf on Big Data
Intelligence and Computing and Cyber Science and
Technology Congress.

[6] Ajay Kumar Sahul, Shivani Dubey, Ashish Kumar Jha,

Ritik Bhargava, Priyanshu Priya, Rupa Kumari “Voice
Assistant Using Artificial Intelligence”

[7] Nallamothu M., & Mukkamala R. “A study on Speech

Recognition and Desktop Assistant using Python”
Journal of Emerging Technologies and Innovative
Research 2019.

© 2024, IRJET | Impact Factor value: 8.226 | ISO 9001:2008 Certified Journal | Page 2061