International Journal of Scientific Research in Engineering and Management (IJSREM)

Volume: 09 Issue: 04 | April - 2025 SJIF Rating: 8.586 ISSN: 2582-3930

An AI-Powered Desktop Voice Assistant for Windows

Dilkash D. Mukadam1, Fareeha M. Majgaonkar2, Swarali K. Khot3, Trupti D. Gite 4,

Girish G. Bhide5
1, 2, 3, 4
Student /Department of Information Technology,
Finolex Academy of Management and Technology, Ratnagiri
5
Faculty /Department of Electronics and Telecommunication Engineering,
Finolex Academy of Management and Technology, Ratnagiri

---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - This work introduces a Windows-compatible 1. INTRODUCTION
AI assistant that executes tasks via voice commands.
Our architecture implements two key AI components: What is voice assistant and how it works. Many of us
(1) real-time language interpretation (2) adaptive might have already known about this voice assistant and
command recognition. This assistant will enable users to we use this in our day-to-day life. A voice assistant is a
perform a range of tasks, such as managing files, digital assistant that uses voice recognition, language
opening and closing different applications, searching the processing algorithms, and voice synthesis to listen to
information on web, setting reminders, creating a note, specific voice commands and return relevant information
opening and managing the files, opening various AI or perform specific functions as requested by the user.
tools, managing the drive, opening the calculator and do
These personal assistants can be easily configured to
some basic calculations and even controlling system
perform many of your regular tasks by simply giving
settings all through voice commands. The voice assistant
voice commands. The Most famous application of
leverages advanced speech recognition technology to
iPhone is “SIRI” which helps the end user to
accurately interpret and process human voice
communicate end user mobile with voice and it also
commands. Upon receiving user input, the system
responds to the voice commands of the user. Same kind
dynamically executes requested tasks or retrieves
of application is also developed by the Google that is
relevant information. Key capabilities include real-time
“Google Voice Search”
voice-to-text conversion, context-aware natural language
understanding, and seamless integration with third-party which is used for in Android Phones. But this
APIs to extend its functionality. Designed for universal Application mostly works on the desktop.
accessibility, the intuitive interface ensures effortless It accepts spoken commands via microphone array or
interaction for users of all technical proficiencies. The typed queries through the GUI. Modern AI voice
user interface is designed to be intuitive and user- assistants boost productivity by enabling hands-free PC
friendly, providing a seamless experience for both control, reducing task time by 30-40%. Through AI-
novice and experienced users. By creating a personal powered voice automation we save our time and
desktop assistant that combines convenience, contribute in other works.
automation, and personalized features, this project aims
to enhance users' productivity and efficiency in their 1.1 LITERATURE REVIEW
day-to-day computer tasks.
This study examines the acceptability of voice-
activated personal assistants (VAPA) in public areas and
Key Words: Desktop, Voice-based, Integration, emphasizes the concerns of users related to privacy,
Language, Voice Assistant, Action, Response, GUI social norms and usability. Research emphasizes the
importance of VAP design, which is in line with the
expectation of users and public labels. It also discusses
the potential of VAP to increase the interaction with the
human computer in a shared environment. These findings

© 2025, IJSREM | www.ijsrem.com DOI: 10.55041/IJSREM45648 | Page 1

 International Journal of Scientific Research in Engineering and Management (IJSREM)
Volume: 09 Issue: 04 | April - 2025 SJIF Rating: 8.586 ISSN: 2582-3930

provide valuable knowledge for developers aimed at availability and productivity. It also deals with
improving public usability of voice assistants [1]. restrictions on internet connection and language support
[6].
This article represents an AI -based voice assistant
to improve users' interaction through natural language The study discusses the integration of a
and machine learning. The authors discuss the technical combination of NLP techniques and machine learning
architecture and functionality of the system and models, and voice recognition technologies to create
emphasize its potential applications in intelligent responsive and context-aware personal assistants. The
environments. The study emphasizes the growing role of authors examine various existing intelligent assistants,
AI in increasing more intuitive and efficient assistants. It such as Apple’s Siri, Google Assistant, and Amazon
also deals with challenges such as user accuracy and Alexa, comparing their functionalities, architectures, and
adaptability [2]. limitations. Additionally, the paper highlights challenges
like data privacy, user adaptation, and multilingual
This research focuses on the development of a AI-
support, while also speculating on future advancements,
powered voice assistant developed with Python’s speech
including enhanced contextual understanding and
recognition libraries (Speech Recognition, PyAudio) and
emotion recognition. This work contributes to the
NLP frameworks (NLTK, spaCy) for intent processing
broader discourse on human-computer interaction and
and describes in detail the implementation of speech
AI-driven automation. [7].
recognition and text functions on speech. The post
emphasizes the simplicity and efficiency of Python The study explores the integration of automatic
libraries, such as speech recognition and pyaudio in speech recognition (ASR) paired with NLP, and
building voice systems. It also discusses potential automation techniques to enhance user interaction with
applications in automation and user. The study serves as computers. The authors describe the development
a practical guide for developers who are interested in process, including the use of Python libraries such as
creating voice solutions [3]. Speech Recognition, pyttsx3, and nltk to enable voice
command execution. The paper highlights various
This work examines the creation of A Python-
functionalities of the assistant, such as web searches,
based ASR system leveraging Speech Recognition and
email automation, and smart home control. Additionally,
PyAudio and emphasizes the integration of libraries such
it discusses challenges related to accuracy, response
as PytSX3 and speech recognition. The authors discuss
time, and security, providing insights into future
the ability of the system to perform tasks such as voice
enhancements for more efficient and intelligent virtual
commands and text conversions. The contribution
assistants [8].
emphasizes the growing availability of speech
technology for developers. It also emphasizes the 1.2 TECHNOLOGIES USED:
potential of such systems in increasing productivity and
user experience [4]. a. Python: Python is a well-liked programming language
at a high level that is recognized for its
This study represents the development of AI
straightforwardness, comprehensibility, and user-
assistant on the area using Python focusing on its ability
friendliness. Python 3.10.0 is being utilized in the
to perform tasks through voice commands. The authors
development of this AI powered voice assistant.
discuss the integration of AI processing and natural
b. Visual Studio Code: Microsoft's Visual Studio Code
language to improve the user interaction. The
(VS Code), the extensible code editor, provides
contribution emphasizes the potential of the system in
comprehensive
automating routine tasks and improving efficiency. It
language support including Python, JavaScript, and
also deals with challenges such as accuracy and response
C++ through its modular extension system.
time [5].
2. METHODOLOGY
This research focuses on designing a voice personal
assistant for PCS using Python and emphasizes its
There are three modules in this assistant. The first
application in simplifying user interactions. The authors
step is for the assistant to bring the voice user input.
discuss the use of Python libraries to achieve speech
Second, analysing the user's input and translate it into the
recognition and automation of tasks. The contribution
appropriate intention and function. The third is an
emphasizes the potential of the system to increase
assistant who provides the user with the result all the
© 2025, IJSREM | www.ijsrem.com DOI: 10.55041/IJSREM45648 | Page 2
 International Journal of Scientific Research in Engineering and Management (IJSREM)
Volume: 09 Issue: 04 | April - 2025 SJIF Rating: 8.586 ISSN: 2582-3930

time through speech. The assistant first begins to receive

human entry. When the assistant receives an input, he
3. SYSTEM ARCHITECTURE
transforms an analog voice input into digital text. If the
assistant is unable to turn the voice into text, he will ask
the user to enter again. After transformation, it begins to Virtual attendees use automated language
process input and mapping into a certain function. The understanding (NLP) to match the user's text or voice
output will then be provided by the user with a voice entry with executable commands. When a user calls a
order. This basic workflow of model is shown in Fig.1. virtual assistant designed to accomplish specified tasks,
Users require or query to get cuttings into segregated then the audio representation of natural speech becomes
commands, making it easier to recognize our voice an executable command or digital data that the software
assistant on the desktop. can analyze. Then, these data are compared to software
data to find an adequate response. The virtual wizard is
used to execute machines in the user provided
commands. For the development of the assistant, we use
Python installers packages as voice recognition, GTTS,
Pip Win, etc. Speech-to-text systems process audio
inputs through feature extraction and language modeling.
This is commonly used in voice assistants such as Alexa,
Siri, etc. Python provides an API called voice recognition
to allow the user to convert the voice or audio command
into text for subsequent processing. As illustrated in
Figure 1, users give the command to interaction entities
Figure 1: Basic workflow of model such as laptop or PC interaction entities listen to the
command and recognize it. For a subsequent analysis, the
• In comparison with other questions, our assistant process compares this command with the cloud of the
searches inside the command list . data that is already stored. If the application coincides
• The voice assistant receives these orders via the with the cloud data, the outputs generated in the form of
command list. text and voice, if not, will give a related message. Look
for the function or logic that will be executed according
• Once the voice assistant has taken or received an to the application and send the output of the backend
order, it immediately sets the appropriate measures to process in response.
submit.
• If the user's query is not understandable, then the 3.1 ACTIVITY DIAGRAM
voice assistant asks for clarification before continuing.
• Especially, the voice assistant detects what we want The overall workflow of the proposed work is
to get. explained with the help of the activity diagram given in
Fig.2.
• When the voice assistant recognizes the order and
finds that it can continue, it provides the person or the
user with the necessary information.
For example, when a person says, open WhatsApp or
Wikipedia: “The voice assistant listens to the order and
takes the appropriate event such as opening the site.
After completing his speaking, the voice assistant stops
for a brief 2-3 second interval to ensure that he captures
the entire application and then searches her database for
the investigation to provide the appropriate result.

© 2025, IJSREM | www.ijsrem.com DOI: 10.55041/IJSREM45648 | Page 3

 International Journal of Scientific Research in Engineering and Management (IJSREM)
Volume: 09 Issue: 04 | April - 2025 SJIF Rating: 8.586 ISSN: 2582-3930

Software applications, performing system-level

operations, retrieving information from databases or
external
sources, or providing a response to the user.
• Generate Response: If the executed action
necessitates a response, the system generates a suitable
response,
either as text or synthesized speech.
• Output Response: The system presents the response
to the user, which could involve converting it into
speech, displaying it on a screen, or both.
• End: The activity diagram end when the virtual
assistant is terminated or deactivated.

3.2 USE CASES

The interaction of the user with the AI powered voice

assistant for performing various tasks on the desktop are
explained with the help of use case diagram given in Fig.
Fig. 2. Activity Diagram 3.

Initiation: The activity diagram begins with the

initialization of the voice-based virtual assistant.
• Listen for Voice Command: The virtual assistant
remains in a continuous listening state, awaiting voice
commands from the user.
• Perform Speech Recognition: The system applies
speech recognition to the user’s voice command,
converting
it into text or a structured command.
• Process Command: The system analyses the
recognized command, aiming to understand its intention
and
extract relevant information.
• Determine Action: Based on the recognized
command and extracted details, the system decides the
appropriate
Fig. 3. Use case Diagram
action or task to be executed.
• Execute Action: The system carries out the
The virtual assistant system is represented by a
determined action, which may involve interacting with
rectangle and consists of several use cases.
others.

© 2025, IJSREM | www.ijsrem.com DOI: 10.55041/IJSREM45648 | Page 4

 International Journal of Scientific Research in Engineering and Management (IJSREM)
Volume: 09 Issue: 04 | April - 2025 SJIF Rating: 8.586 ISSN: 2582-3930

• The user activates the assistant by initiating the” f"Temp_Notepad_File_{len(open_files) + 1}.txt")

Activate Assistant” use case. subprocess.Popen(["notepad.exe",
• The user performs voice commands by engaging in temp_file_path])
the” Perform Voice Command” use case. open_files.append(temp_file_path)

• The user has the option to configure the assistant by elif "open new tab" in response:
participating in the” Configure Assistant” use case. speak("Opening a new tab in Notepad.")
• The” Perform Voice Command” use case leads to temp_file_path =
the” Process Command” use case, where the assistant os.path.join(os.environ["USERPROFILE"], "Desktop",
f"Temp_Notepad_File_{len(open_files) + 1}.txt")
analyses and comprehends the received voice
subprocess.Popen(["notepad.exe",
command.
temp_file_path])
• Upon understanding the command, the assistant open_files.append(temp_file_path)
proceed to the” Execute Action” use case, where it
performs the appropriate action. elif "previous" in response or "saved" in response:
speak("What is the name of the file you want to
Once the action is executed, the assistant enters the”
open?")
Provide Response” use case, generating a response for
file_name = takeCommand().strip()
the user.
• The user can interact with the output or response if file_name: # Only proceed if a valid file name is
through the” Interact with Output” use case. given
• The loop indicates that the user can continue search_directories = {
performing voice commands and interacting with the "Desktop":
assistant as needed. os.path.join(os.environ["USERPROFILE"], "Desktop"),
"Documents":
os.path.join(os.environ["USERPROFILE"],
3.3 SAMPLE CODE "Documents")
}
Several activities based on the desktop can be
performed using this AI powered voice assistant. For found_file = None
every such activities separate codes are developed. A for location, directory in search_directories.items():
sample code out of these codes is given in Table 1. file_path = os.path.join(directory, file_name + ".txt")
if os.path.exists(file_path):
Table 1. Sample code for opening a Notepad speak(f"Opening the file {file_name} saved in
def open_notepad(): {location}.")
"""Opens Notepad and listens for user commands to subprocess.Popen(["notepad.exe",
open new tabs, write, or manage files.""" file_path])
try: open_files.append(file_path)
open_files = [] # List to track opened files found_file = file_path
break
while True:
speak("Do you want to open a new file, a previously if not found_file:
saved file, or open a new Notepad window?") speak(f"Sorry, I couldn't find a file named
response = takeCommand().lower() {file_name} in your Documents or Desktop.")
else:
if "new file" in response or "open new window" in speak("I didn't understand. Opening a new Notepad
response: file by default.")
speak("Opening a new Notepad file.") temp_file_path =
temp_file_path = os.path.join(os.environ["USERPROFILE"], "Desktop",
os.path.join(os.environ["USERPROFILE"], "Desktop", f"Temp_Notepad_File_{len(open_files) + 1}.txt")

© 2025, IJSREM | www.ijsrem.com DOI: 10.55041/IJSREM45648 | Page 5

 International Journal of Scientific Research in Engineering and Management (IJSREM)
Volume: 09 Issue: 04 | April - 2025 SJIF Rating: 8.586 ISSN: 2582-3930

subprocess.Popen(["notepad.exe", file_name + ".txt")

temp_file_path]) location = "Desktop"
open_files.append(temp_file_path) elif "documents" in location_command:
saved_file_path =
break # Proceed to listening mode os.path.join(os.environ["USERPROFILE"],
"Documents", file_name + ".txt")
# **Wait for additional commands** location = "Documents"
while True: else: speak("I didn't understand. Saving on
print("Waiting for a command while Notepad is Desktop by default.")
open...") saved_file_path =
command = takeCommand().lower() os.path.join(os.environ["USERPROFILE"], "Desktop",
file_name + ".txt")
if "open new tab" in command: location = "Desktop"
speak("Opening a new Notepad tab.")
temp_file_path = # Rename the last opened temp file
os.path.join(os.environ["USERPROFILE"], "Desktop", if open_files:
f"Temp_Notepad_File_{len(open_files) + 1}.txt") os.rename(open_files[-1], saved_file_path)
subprocess.Popen(["notepad.exe", speak(f"File '{file_name}' has been saved on
temp_file_path]) {location}.")
open_files.append(temp_file_path) open_files[-1] = saved_file_path

elif "write in notepad" in command: elif "save" in command:

speak("Please dictate the content you want to add.") speak("Saving changes in the opened files.")
content = takeCommand()
elif "close notepad" in command:
temp_file_path = speak("Closing all Notepad windows.")
os.path.join(os.environ["USERPROFILE"], "Desktop", subprocess.call(["taskkill", "/F", "/IM",
f"Temp_Notepad_File_{len(open_files) + 1}.txt") "notepad.exe"], stdout=subprocess.DEVNULL,
with open(temp_file_path, "w") as f: stderr=subprocess.DEVNULL)
f.write(content) break # Exit loop once Notepad is closed

subprocess.Popen(["notepad.exe", time.sleep(1) # Small delay before checking the next

temp_file_path]) command
speak("Your dictated text is saved and opened in
Notepad.") return True, open_files # Return the list of opened
open_files.append(temp_file_path) Notepad files

elif "save file" in command: except Exception as e:

speak("What should be the name of the file?") speak("An error occurred while opening Notepad.")
file_name = takeCommand().strip() print(f"Error: {e}")
if not file_name: return False, None
file_name = "Untitled"
For access to additional code and resources,
speak("Where do you want to save the file? Say please refer to the following Drive link.
'Desktop' or 'Documents'.")
location_command = takeCommand() https://drive.google.com/drive/folders/1LUe8tC4
1OwuzzpFifu7TDmw9S21Ya6NJ
if "desktop" in location_command:
saved_file_path =
os.path.join(os.environ["USERPROFILE"], "Desktop",

© 2025, IJSREM | www.ijsrem.com DOI: 10.55041/IJSREM45648 | Page 6

 International Journal of Scientific Research in Engineering and Management (IJSREM)
Volume: 09 Issue: 04 | April - 2025 SJIF Rating: 8.586 ISSN: 2582-3930

4. RESULTS & FUTURE SCOPE Opening Apps Can open apps Can open apps
like Notepad, but fails in 20%
4.1 RESULTS Calculator, of cases (e.g.,
Chrome, etc. misinterprets the
The GUI created and used for this work is shown app name).
in Fig. 4. All the use cases were thoroughly tested for
functionality. The results are encouraging. All these Opening Can open Cannot handle
results were used then for comparison with existing voice Notepad and Notepad with file-saving
assistants. One of the results, which is response to “Open Asking to Save multiple tabs commands.
Notepad” voice command is shown as a sample in Fig. 5 when asked to Fails to
open, ask to recognize
save the file, "save" or "file
and specify the name"
file name. commands.

Opening Can open Can perform

Calculator and Calculator and calculations but
Performing perform cannot open
Calculations calculations Calculator

Increasing and Can increase, Can adjust

Decreasing decrease, and volume on
Volume mute volume. supported
devices (e.g.,
Siri for iPhone,
Fig. 4. GUI of the proposed work Alexa for Echo).

Video Playback Can play, pause, Can control

Controls stop, fast- playback on
forward, rewind, compatible
and resume services
videos.

Google Drive Can open Can assist in

Access Google Drive searching
instantly. Google Drive
Fig. 5. Response for the voice command “Open files but not
notepad” open via voice
(Copilot)
The comparison of proposed work with existing
voice assistants against various functionalities is shown Closing Can close apps Cannot close
in Table 2. Applications like Notepad, apps
Calculator,
Chrome, etc.
Table 2. Proposed work against other voice assistants
– A Comparison File Can search for Can search for
Management and open saved files in
Functionality Proposed Work Other Voice (Searching & Notepad files, OneDrive but
Assistants Opening Files) Word Files, etc. not system-wide
(Google
Assistant, Siri, File Can copy, move, No support for
Alexa) Management and delete files file management

© 2025, IJSREM | www.ijsrem.com DOI: 10.55041/IJSREM45648 | Page 7

 International Journal of Scientific Research in Engineering and Management (IJSREM)
Volume: 09 Issue: 04 | April - 2025 SJIF Rating: 8.586 ISSN: 2582-3930

(Copy, Move, with voice operations. The graphical representation of the above
Delete Files) commands. comparison is shown in Fig. 6. The overall efficiency of
the proposed work comes out to be 92.33%, which is
AI-Powered Includes AI- Available in very good against 52.78% efficiency of the existing
Summarization powered Copilot for voice assistants [9, 10].
summarization Microsoft 365
and web search (Word, Outlook)
functionality via but not system-
AI wide

Web Search via Can search the Available via

AI web with AI Copilot (Bing
AI), but no
direct voice
control

Offline Works offline WSR works

Capabilities for app control, offline but lacks
calculations, AI
and file Fig. 6. Comparison of accuracy of functionalities
management

Shutdown and Can restart and Not available in

4.2 FUTURE SCOPE
Restart Pc shut down PC Copilot or other
via voice assistants
Looking ahead, there’s a lot of exciting potential
commands for this work. By using machine learning, the assistant
Take, Read and Can take notes, Not supported. could get even smarter, understanding conversations
Delete notes read saved Assistants like better and responding more naturally. It could also be
notes, and delete Siri and Alexa upgraded with a text-based chatbot feature, giving users
specific notes do not offer the choice to either type or talk. Adding support for
via voice direct note multiple languages would make it even more accessible
commands. deletion. to people around the world. Over time, the assistant
could learn about user preferences to offer more
Grammar Can check and Not available in personalized and helpful experience. It could also take
checking correct grammar voice assistants over everyday tasks like managing emails, setting up
in Word or Copilot calendar events, and taking notes. With web scraping, it
documents. (requires could bring live updates like news and weather directly
manual input). to the user. Connecting it to smart home devices would
turn it into a powerful tool for home automation. Plus,
Taking Can take Not available in making it available through Windows, Mac, Linux, and
Screenshot screenshots via other voice mobile devices would ensure it’s always within reach. To
voice command assistants. top it off, security could be strengthened with voice
and save them. authentication, keeping user data safe and ensuring a
more personal touch.
Graphical Custom GUI for Primarily voice-
Control System user interaction, only, no
5. CONCLUSION
(GUI) displaying customizable
commands, GUI
This proposed work presents an intelligent voice
outputs, and
interface developed for the Windows desktop platform,
controls
integrating AI to enhance user interaction through speech
recognition and automation. The assistant efficiently

© 2025, IJSREM | www.ijsrem.com DOI: 10.55041/IJSREM45648 | Page 8

 International Journal of Scientific Research in Engineering and Management (IJSREM)
Volume: 09 Issue: 04 | April - 2025 SJIF Rating: 8.586 ISSN: 2582-3930

performs tasks such as opening and closing applications, in Science, Communication and Technology
file management, mathematical calculations, volume (IJARSCT), Volume 6, Issue 2, June 2021. ISSN
control, and video playback using voice commands. (Online): 2581-9429
[6] V Geetha & Gomathy, C K & Kottamasu, Manasa
A comparative analysis with an existing project & Kumar, Nukala. (2021). The Voice Enabled
highlights significant improvements in accuracy and Personal Assistant for Pc using Python. International
performance. Our assistant achieves an overall accuracy Journal of Engineering and Advanced Technology.
of 92.33%, compared to 52.78% for the existing system. 10. 162-165. 10.35940/ijeat.D2425.0410421
Notable enhancements include 100% accuracy in [7] Aditya Sinha, Gargi Garg, GouravRajwani, Shimona
opening applications, 96% accuracy in file searching, Tayal, “Intelligent Personal Assistant”, International
and 92% accuracy in mathematical calculations, Journal of Informative Futuristic Research, Volume.
outperforming previous models in every key 4, Issue 8, April 2017.
functionality.
[8] Vadaboyina Appalaraju, V Rajesh, K Saikumar, P.
Sabitha” Design and Development of Intelligent
The results indicate the proposed assistant offers
Voice Personal Assistant using Python” 2021 3rd
a more reliable and user-friendly experience, with
International Conference on Advances in
potential applications in productivity, automation, and
Computing, Communica- tion Control and
accessibility. Future work could explore AI-driven
Networking (ICACCCN)
enhancements for better contextual understanding and
expanded functionality. [9] Silky Sharma, Prof.(Dr.) Gurinder Singh,
“Comparison of Voice Based Virtual Assistants
REFERENCES fostering Indian Higher Education – A Technical
Perspective”, 2021 International Conference on
[1] Easwara Moorthy, Aarthi & Vu, Kim-Phuong, Technological Advancements and Innovations
“Voice Activated Personal Assistant: Acceptability (ICTAI), November 2021, DOI:
of Use in the Public Space” HIMI 2014. Lecture 10.1109/ICTAI53825.2021.9673307
Notes in Computer Science, vol 8522. Springer, pp. [10] Andreas M. Klein, Maria Rauschenberger, Jorg
324-334, 10.1007/978- 3-319-07863-2_32.J. Clerk Thomaschewski, and Maria Jos´e Escalona,
Maxwell, A Treatise on Electricity and Magnetism, “Comparing Voice Assistant Risks and Potential
3rd ed., vol. 2. Oxford: Clarendon, 1892, pp.68–73. with Technology-Based Users: A Study from
[2] Subhash, P. N. Srivatsa, S. Siddesh, A. Ullas and B. Germany and Spain”, Journal of Web Engineering,
Santhosh, "Artificial Intelligence-based Voice Vol. 20 7, pp 1991–2016, doi: 10.13052/jwe1540-
Assistant," 2020 Fourth World Conference on Smart 9589.2071, November 2021.
Trends in Systems, Security and Sustainability
(WorldS4), 2020, pp. 593-596, doi:
10.1109/WorldS450073.2020.9210344K. Elissa,
“Title of paper if known,” unpublished.
[3] Harshit Agrawal, Nivedita Singh, Gaurav Kumar,
Dr. Diwakar Yagyasen, Mr. Surya Vikram Singh
“Voice Assistant Using Python” 2021, IJIRT
Volume 8 Issue 2, ISSN: 2349-6002, pp.419-423.
[4] Mrs.A.M.Sermakani, J.Monisha, G.Shrisha,
G.Sumisha, “Creating Desktop Speech
Recognization Using Python Programming.”
IJARCCE, Vol. 10, Issue 3, March 2021, ISSN
(Online), pp.129-134
[5] Abeed Sayyed, AshpakShaikh,
AshishSancheti,Swikar Sangamnere, Prof. Jayant H
Bhangale. “Desktop Assistant AI Using Python”
(2021) International Journal of Advanced Research
© 2025, IJSREM | www.ijsrem.com DOI: 10.55041/IJSREM45648 | Page 9