1

The Way To Make Blind People Use E-Mail

System: Voice Based E-Mail Generating System
Using Artificial Intelligence
Vibhor Sharma , Gold Medalist B.E., cr69

Abstract—In the present scenario, communication technology Communication technologies are significant these days for the
is essential for connecting with each other. This paper proposes betterment of social and personal interaction. The combination
a voice-based email system using artificial intelligence to assist of technologies with the internet makes communication easy.
new users and physically impaired individuals in effective com-
munication. The system relies on mouse actions and voice con- However, the person who is physically challenged suffered
version, making it accessible without the need for prior practices. a lot to utilize this technology due to visual and physical
Through speech recognition and text-to-speech, even non-literate difficulties. There are many technologies advancements have
individuals can send emails. The design emphasizes responsive come though it is not possible to use like normal users. This
voice interaction for a user-friendly experience, allowing quick paper aims at creating an email system that helps even new
email sending and comprehensive system functionality.
users or physically impaired people to use the system for
Index Terms—Voice-based communication, Artificial Intelli- communication without any previous practices. There is no
gence, Email system, Accessibility, Responsive interaction. use of keywords, only with the help of mouse actions and
voice conversion the email system works. The person who
I. I NTRODUCTION is not literate can also send emails as it is based on speech
recognition and text to speech. The system is completely
In the modern era, where communication technology plays a
based on responsive voice interaction to utilize the technology
pivotal role in connecting individuals, its accessibility remains
easy and hassle free manner. The system is well designed to
a challenge for those facing physical constraints. This research
send the mails quickly. There are all the options available to
introduces a groundbreaking solution to bridge this gap: a
send emails and perform all the functions for the email system.
Voice-Based E-Mail Generating System utilizing Artificial
Intelligence (AI). Communication technologies are essential
The ”Discovery of Activities’ Actor Perspective from
for social and personal interactions, yet individuals with
Emails based on Speech Acts Detection” [2] by Marwa
physical challenges struggle to harness their full potential
ELLEUCH ,Oumaima ALAOUI ISMAILI,Nassim LAGA
due to visual and physical limitations. This paper addresses
Nour ASSY , Walid GAALOUL. Process mining aims at
this issue by proposing an innovative email system that pri-
discovering different perspectives of business processes (BP)
oritizes inclusivity. Unlike conventional systems, this voice-
from event logs generated by BP management systems.
based approach relies on mouse actions and AI-driven voice
However, BP can be entirely or partially performed outside
conversion, enabling even new users and physically impaired
such systems. Emails are widely used as an alternative tool
individuals to engage in effective communication without the
to collaboratively perform BP tasks. Recently, there have
need for extensive prior practices.
been several initiatives to extend the scope of BP mining to
The system leverages advanced technologies such as speech
consider email logs. However, most of them have been mainly
recognition and text-to-speech, making it accessible to non-
focused on discovering BP activity perspective neglecting
literate individuals. The emphasis on responsive voice inter-
other equally important knowledge. Actor perspective is
action ensures a user-friendly experience, facilitating quick
one of the important knowledge that can ensure more
email sending and comprehensive system functionality. The
understanding regarding individuals acting for performing BP
introduction explores the background of communication tech-
activities. Mining such perspective from emails logs provides
nologies, delving into existing challenges and the need for
additional information about the precise contribution of actors
inclusive solutions. As we delve into the methodology, re-
in the execution of BP activities. Such information is not
sults, and conclusion, this research contributes significantly to
limited to executing activities but also refers to requesting,
the advancement of accessible communication technologies,
informing, planning or observing activities’ execution. This
marking a step towards a more inclusive digital landscape. b
paper first formalizes the knowledge we may discover from
emails related to actors perspectives. Then, it introduces an
II. L ITERATURE S URVEY approach based on speech act detection from textual content
The ”Voice Email Based On SMTP For Physically of emails for discovering such knowledge.Our approach is
Handicapped” [1] by Sunny Kumar, Ms. Yogitha , Ms. validated using a public email dataset. Our results are publicly
R. Aishwarya. In the present scenario, everybody needs provided to be a first step towards ensuring reproductibility
communication technology to connect with each other. in the studied area.,
 2

structured methodology to ensure the effective processing of

The ”Email classification via intention-based segmentation” voice commands and seamless email composition. The follow-
[3] by Sanjay Kumar Sonbhadra, Sonali Agarwal, Mohammad ing procedural steps delineate the methodology employed for
Syafrullah, Krisna Adiyarta. Email is the most popular way system development:
of personal and official communication among people and
organizations. Due to untrusted virtual environment, email A. Voice Input Processing
systems may face frequent attacks like malware, spamming, • Implement a robust Speech Recognition Module using
social engineering, etc. Spamming is the most common advanced speech recognition libraries and algorithms,
malicious activity, where unsolicited emails are sent in bulk, such as the Google Speech-to-Text API.
and these spam emails can be the source of malware, waste • Define internal data structures governing the structure
resources, hence degrade the productivity. In spam filter and processing mechanisms for voice inputs within the
development, the most important challenge is to find the Speech Recognition Module.
correlation between the nature of spam and the interest of the
users because the interests of users are dynamic. This paper
B. Natural Language Processing (NLP)
proposes a novel dynamic spam filter model that considers
the changes in the interests of users with time while handling • Employ the NLP Module to convert voice-generated text
the spam activities. It uses intention-based segmentation to into a machine-understandable format.
compare different segments of text documents instead of • Specify the data structure within the NLP Module, out-
comparing them as a whole. The proposed spam filter is lining the representation of parsed data.
a multi-tier approach where initially, the email content is
divided into segments with the help of part of speech (POS) C. Email Composition
tagging based on voices and tenses. Further, the segments are • Design an Email Composition Component that extracts
clustered using hierarchical clustering and compared using pertinent information from the NLP module for the
the vector space model. In the third stage, concept drift is composition of email messages.
detected in the clusters to identify the change in the interest • Provide detailed descriptions of functionalities and inter-
of the user. Later, the classification of ham emails into various connections within the Email Composition and Sending
categories is done in the last stage. For experiments Enron Component.
dataset is used and the obtained results are promising.
D. Recipient Management
The ”A Review on Voice based E-Mail System for
• Implement a function within the system to process voice
Blind” [4] by Paulus A. Tiwari,Pratiksha Zodawan,Harsha
commands for managing email recipients.
P. Nimkar,Trishna Rotke,Priya G. Wanjari,Umesh Samarth.
• Define internal data structures governing recipient infor-
Due to its simplicity and accessibility, Internet is widely used
mation.
in almost all the communication applications. In the recent
times, number of application based on internet have been
E. Attachment Handling
developed to make the communication as a more reliable and
efficient in nature. Out of this numerous applications, E-mail • Develop a module facilitating the attachment of files or
is the most widely used and reliable way to communicate with documents to emails through voice commands.
each other. The usage of e-mail is quiet easy and lucid for • Specify the data structure for handling attachments within
regular users but when it comes to the user with visual defect, the system.
the system is yet very difficult to use. The current system
is not useful for people with visual defect as the available F. Voice Confirmation
system are based on the visual perceptions. There are huge • Design a Voice Confirmation Component to read back
up gradation in the technologies now a days, especially for composed emails to the user for verification.
the visually challenged people. Still the current emailing • Ensure seamless integration with the Speech Recognition
system is yet not upgraded for the use of visually impaired. and NLP modules.
This arises a significant need to upgrade the existing system
to make it more useful for the visually impaired. Thus, in G. Email Sending
this study we present an email system working on the voice
• Implement the Email Sending Component to transmit
controlling principle for the people with visual impairment to
confirmed emails to specified recipients.
deliver a simple and easy access to the email system. This
• Define the technology stack used for email transmission,
framework will also helpful for the individuals with other
ensuring secure and efficient delivery.
weaknesses alongside the visually impaired individuals.

H. Technology Stack
III. M ETHODOLOGY
1) Speech Recognition Module: Utilizes cutting-edge
The development of the proposed Voice-Based E-Mail Gen- speech recognition libraries and algorithms, such as the Google
erating System using Artificial Intelligence (AI) adheres to a Speech-to-Text API.
 3

IV. R ESULT
• Enhanced Accessibility:
– The implemented system successfully addresses the
accessibility challenges faced by individuals with
physical impairments, providing them with a viable
and efficient means of engaging in email communi-
cation.
• Efficient Voice Input Processing:
– The Speech Recognition Module demonstrated com-
mendable accuracy and processing speed, effectively
capturing and processing user voice inputs.
• Robust Natural Language Processing:
– The NLP Module proved effective in converting
voice-generated text into a machine-understandable
format, showcasing its capability in interpreting user
intent and extracting relevant information.
• User-Friendly Email Composition and Sending:
– The Email Composition and Sending Component
exhibited reliable performance, enabling users to
compose and send emails seamlessly through voice
commands.
• Effective Recipient Management and Attachment Han-
Fig. 1: System Architecture dling:
– The system’s functionalities for managing email
recipients and handling attachments through voice
2) NLP Module: Incorporates Natural Language Processing commands contribute to a comprehensive and effi-
libraries, such as NLTK or spaCy, for effective language cient email communication experience.
understanding. • User Verification through Voice Confirmation:
3) Email Composition and Sending Component: Developed – The Voice Confirmation Component ensures user
using programming languages, e.g., Python, with libraries for verification by reading back composed emails,
email handling, such as smtplib. thereby enhancing the accuracy and correctness of
This methodology ensures the seamless integration of voice email content.
input processing, NLP analysis, and email composition, pro- • Positive User Experience:
viding an inclusive and user-friendly platform for individuals – User feedback indicates a positive experience with
with varying abilities. The chosen technology stack guarantees the system, emphasizing its ease of use, intuitiveness,
the reliability and effectiveness of each system component. and overall user satisfaction.
• Future Directions:
– While the current implementation demonstrates sig-
nificant advancements, there is room for further
enhancements and expansions. Future research could
explore additional features, improve system accu-
racy, and extend accessibility to a broader user base.

V. C ONCLUSION
In conclusion, the Voice-Based E-Mail Generating System
leveraging Artificial Intelligence offers a groundbreaking solu-
tion for inclusive communication technology. This project suc-
cessfully addresses accessibility challenges faced by individu-
als with physical impairments, providing an efficient means for
email communication through accurate voice input processing
and robust natural language understanding. The system’s user-
friendly email composition and sending capabilities, coupled
Fig. 2: Use Case diagram with positive user feedback, affirm its effectiveness. The
comprehensive functionality, including recipient management,
attachment handling, and voice confirmation, contributes to a
 4

holistic and enriched user experience. This research establishes

a significant milestone in fostering accessible and inclusive
communication technologies, with potential applications for
diverse user groups. Future directions may focus on further
enhancements and broader accessibility, extending the positive
impact of this innovative system.

ACKNOWLEDGMENT
This research project has been made possible through the
support and contributions of various sources. We extend our
gratitude to the authors of the referenced research papers,
whose work has been instrumental in shaping and guiding our
project.

R EFERENCES
[1] Voice Email Based On SMTP For Physically Handi-capped
[2] Discovery of Activities’ Actor Perspective from Emails based on Speech
Acts Detection
[3] Email classification via intention-based segmentation
[4] A Review on Voice based E-Mail System for Blind