Voice Based Email Generating System Using AI
CHAPTER 1
INTRODUCTION
This foundation is designed to assist the colorblind or partially blind as well as the
untrained. Our task seems to be an applications for a customer that appears to be disabled
or an illiterate customer but wants to use online services like that of a regular citizen. This
tool will help overcome a few difficulties that would have previously were encountered
while allowing access to communications for those who are physically weaker than
others. The simplest affirmative interface shortcuts might be used with this device.
Managing the use of voice awareness might definitely be a successful method for postal
devices for the blind. With this, we put up our challenge to make it possible for those who
are obstructed to easily get their messages in a more informative manner. Due to the
improvement in cellular telephones, several engineering setups got made so that those
who were physically weaker could employ them and profit from them.
The very last relevant message material might be saved in a report. This is
currently being developed using the Velvet Ide on a Mobile platform. Our voice-to-
literary device recognizes sound right away and converts it entirely on print material. By
giving customers a choice of options for metadata, it usually takes either or two large
concepts to finish. Dialogue translator is effectively used to increase the computer's
accessibility to reality selections for customers who are legally blind or severely disabled.
In today's fast-paced world, where communication is paramount, email remains
one of the most ubiquitous forms of interaction. However, traditional methods of
composing emails can be time-consuming and cumbersome, especially for individuals
with busy schedules or limited accessibility to conventional input devices like keyboards.
To address this challenge, the integration of Artificial Intelligence (AI) into email systems
offers a promising solution. Voice-based email generating systems leverage the power of
AI to enable users to compose and send emails effortlessly using voice commands.
This innovative technology not only enhances accessibility for individuals with
disabilities but also streamlines the email composition process for everyone, regardless of
their typing proficiency or time constraints. By harnessing the capabilities of AI, voice-
Department of CS&E, PESITM, Shivamogga. Page 1
Voice Based Email Generating System Using AI
based email systems can understand natural language input, interpret user intent, and
generate accurate and contextually appropriate email content.
1.1 Overview on Artificial Intelligence
Figure 1.1: Artificial intelligence encompasses machine learning and deep learning
The term "artificial intelligence" (AI) was coined in 1956 by John McCarthy
duringa conference held on this subject. Al is the branch of computer science that deals
with designing intelligent computer systems that mimic human intelligence. The ability of
machines to process natural language, to learn, to plan makes it possible for new tasks to
be performed by intelligent systems. The main purpose of Al is to mimic the cognitive
function of human beings and perform activities that would typically be performed by a
human being. Al is stand-alone independent electronic entity that functions much like
human expert. Today, Al is integrated into our daily lives in several forms, such as
personal assistants, automated mass transportation, aviation, computer gaming, facial
recognition at passport control, voice recognition on virtual assistants, driverless cars,
companion robots, etc.. An important feature of AI technology is that is can be added to
existing technologies. Al has benefited many areas such chemistry and medicine, where
Department of CS&E, PESITM, Shivamogga. Page 2
Voice Based Email Generating System Using AI
routine diagnoses can initiated by AI-aided computers. It embraces a wide range of
disciplines such as computer science, engineering, machine learning, chemistry, biology,
physics, astronomy, neuroscience, and social sciences. Al is not a single technology but a
range of computational models and algorithms. The major disciplines in Al include expert
systems, fuzzy logic, and artificial neural networks (ANNs), machine learning, deep
learning, natural language processing, computer vision, and robotics. The various
computer- based tools or technologies that have been used to achieve Al's goals are the
following.
Figure 1.2: An illustration of deep learning with two hidden layers
The above figure 1.2 aims at increasing the capacity of supervised and
unsupervised learning algorithms for solving complex real-world problems by adding
multiple processing layers.
Expert Systems: An expert system (ES) (or knowledge-based system) enables
computers to make decisions by interpreting data and selecting between
alternatives just as a human expert would do. It uses a technique known as rule-
based inference in which rules are used to process data.
Neural Networks: These computer programs identify objects or recognize
patterns after having been trained. Artificial neural networks (ANNs) are parallel
distributed systems consisting of processing units (neurons) that calculate some
mathematical functions. The ANN model represents nonlinear relationships
Department of CS&E, PESITM, Shivamogga. Page 3
Voice Based Email Generating System Using AI
which are directly
Department of CS&E, PESITM, Shivamogga. Page 4
Voice Based Email Generating System Using AI
learned from the data being modeled. Neural networks are being explored for
healthcare applications in imaging and diagnoses, risk analysis, lifestyle
management and monitoring, health information management, and virtual health
assistance.
Natural Language Processors: Computer programs that translate or interpret
language as it is spoken by normal people. Natural language processors (NL)
techniques extract information from unstructured data. NLP targets at extracting
useful information from the narrative text to assist decision making. NLP'includes
applications such as speech recognition, text analysis, translation and other goals
related to language. There are two basic approaches to NLP: statistical and
semantic. Healthcare is the biggest user of the NLP tools. NLP has been used in
the clinical setting for capturing, representing, and utilizing clinical information.
Robots: Computer-based programmable machines that have physical
manipulators and sensors. For example, medical robots can help with surgical
operations, rehabilitation, social interaction, assisted living, etc. Robotic-guidance
is becoming common in spine surgery.
Fuzzy Logic: Reasoning based on imprecise or incomplete information in terms
of a range of values rather than point estimates. Fuzzy logic deals with uncertainty
in knowledge that simulates human reasoning in incomplete or fuzzy data. The
fuzzy model is robust to parameter changes and tolerant to impression.
Machine Learning: Algorithms to make predictions and interpret data and
"learn", without static program instructions. ML is a statistical technique for
fitting models
to data and training models with data. Machine learning (ML) extracts features
from input data by constructing analytical data algorithms and examines the
features to create predictive models. The most common ML algorithms are
supervised learning, unsupervised learning, reinforcement learning, and deep
learning. Machine learning is widely used in human-computer interaction.
Deep Learning: A subset of machine learning built on a deep hierarchy of layers,
with each layer solving different pieces of a complex problem. It aims at
increasing the capacity of supervised and unsupervised learning algorithms for
solving complex real-world problems by adding multiple processing layers. An
illustration of deep learning with two hidden layers is in Figure 1.3. Figure 1.2
shows how Al encompasses machine learning and deep learning.
Department of CS&E, PESITM, Shivamogga. Page 5
Voice Based Email Generating System Using AI
Data Mining: This deals with the discovery of hidden patterns and new
knowledge from large databases. Data mining exhibits a variety of algorithmic tools
such as statistics, regression models, neural networks, fuzzy sets, and evolutionary
models. Each AI tool has its own advantages. Using a combination of these models,
rather than a single model, is recommended. Al technologies are drastically
influencing the retail industry and customer experience. An emerging area of interest
in Al is tomake Al agents cooperate with each other.
1.2 Speech-To-Text Converter
A monologue translation gives the infrastructure the ability to generate outputs.
When anyone speaks into a device and is picked up by a user of the device, the voice is
transformed into written content. My rhetoric system immediately receives and converts
the spoken input to the written text. The technique used to convert uttered utterances into
prepared text is called mouth to letter material transformations. It enables people who are
physically disabled to receive the opportunity to manage the whole infrastructure by
contributing voice without of worrying about using different pathways or showing
assistive technologies. In voice-based email systems, users speak their passwords and
usernames to register into the device while, in addition, they choose which actions they
want to be taken, such as revealing their account, e - mailing, creating new messages.
A Speech-to-Text Converter is a vital tool that transforms spoken language into
written text with remarkable accuracy and efficiency. By leveraging advanced algorithms
rooted in machine learning and neural networks, this software seamlessly transcribes
audio input sourced from microphones, telephony systems, or recorded files. Through the
intricate process of speech recognition, the converter dissects the audio signal into
phonemes or words, matching them against an extensive database of linguistic patterns. it
deciphers the meaning and context of the spoken words, analyzing grammar, syntax,
semantics, and contextual cues. Once interpreted, the converter generates written text
output that faithfully represents the transcribed speech, which can be displayed in real-
time or saved for further processing. To ensure accuracy, the converter incorporates error
correction mechanisms and continuously refines its performance through machine
learning, adapting and improving over time. Integrated into various applications and
services, Speech-to-Text Converters empower users with voice-based interaction, offering
unparalleled convenience, accessibility, and efficiency in communication.
Department of CS&E, PESITM, Shivamogga. Page 6
Voice Based Email Generating System Using AI
1.3 Text-To-Speech Converter
A letter converters makes it possible to get output out from device. When every
action takes place within the device, text composition is the consequence, although it is
useless for those who are physically weak. The intellectual content is subsequently
converted into words and conveyed via them as a result. It is an email (successive)
interactive screen device that can convert word material to voice in audio formats such
printable stuff to playback or textual to wave recorded. It is sometimes referred to as a
letter converting, a type and speak program, or a cultural material client government. As
it does not involve using cursor keys or any other type of surrender exhibit, it is really
useful. The written product function converts the word material it transmits into audio and
is displayed to the user in the voice-based internet foundation when the user provides
instructions to assess actual box starts sending was sent letters.
Text-to-Speech (TTS) converters are essential tools that transform written text into
spoken words. There's a plethora of options available, ranging from web-based services to
standalone software applications. Google Text-to-Speech stands out for its versatility,
offering multiple languages and voices, while Microsoft Azure's Text to Speech service
provides customizable options with natural-sounding synthesis. Amazon Polly, another
cloud-based solution, excels in lifelike speech synthesis across various languages. IBM
Watson's Text to Speech service offers advanced customization for voices and languages.
On the software front, options like NaturalReader, Balabolka, and TextAloud cater to
specific needs, offering features such as voice customization, support for various file
formats, and the ability to save speech as audio files. Whether it's for personal use,
accessibility purposes, or integration into applications, these TTS converters provide
valuable solutions for converting text into spoken language.
1.4 Artificial Intelligence for Speech Recognition
Ai systems (AI) is a technology that is used to construct clever computers and
systems that imitate thinking. Several management architectures, Nlp (Ltp), computational
modeling, and conversation acknowledged are some examples of artificial intelligence
applications that use these technologies. Nlp is the ability to understand and analyses
words and sentences, such as English, by eliminating information underlying expressions
of emotions, relationships, and thoughts.
Department of CS&E, PESITM, Shivamogga. Page 7
Voice Based Email Generating System Using AI
Artificial intelligence (AI) lies at the heart of contemporary speech recognition
systems, furnishing them with the capability to decipher and comprehend human speech.
These systems heavily rely on machine learning algorithms, particularly deep learning
techniques like recurrent neural networks (RNNs) and convolutional neural networks
(CNNs), to process audio data and distill pertinent features for recognition. Furthermore,
AI aids in feature extraction from audio signals, capturing crucial attributes like
spectrograms and Mel-frequency cepstral coefficients (MFCCs) to facilitate the
recognition process. Additionally, AI techniques are instrumental in modeling language
structures and contexts, enabling speech recognition systems to grasp grammar, syntax,
and semantics more accurately. Acoustic modeling, employing methods such as Hidden
Markov Models (HMMs) and deep neural networks (DNNs), further refines recognition
by discerning phonemes, intonation, and speaker characteristics. Leveraging adaptation
and personalization techniques, AI empowers these systems to adjust to diverse speakers,
accents, and environments, bolstering their accuracy and adaptability.
Through continuous learning mechanisms like reinforcement learning, speech
recognition systems refine their models over time, continuously improving their
performance based on user interactions and real-world data. Moreover, integration with
natural language processing (NLP) capabilities augments these systems, enabling them to
comprehend language at a deeper level and execute tasks like transcription, translation,
and voice-controlled commands with greater precision. Overall, AI serves as the
cornerstone of speech recognition, furnishing machines with the capacity to accurately
transcribe and interpret human speech across various contexts and applications.
1.5 NLP (Natural Language Processing)
The interaction across humanity languages and advanced analytics is the focus of
Nlp (Np). In addition to being within the purview of computer science and ai, Lpc is an
important component of supercomputing genealogy. Logical, morphological, emotional
research, vectorization, tokenizing, normalization, and other processes are included.
Natural Language Processing (NLP) stands as a cornerstone in the realm of
artificial intelligence (AI), dedicated to empowering computers with the ability to
comprehend, interpret, and generate human language in a manner that aligns with context
and meaning. Within NLP lie a multitude of tasks and methodologies aimed at processing
and analyzing natural language data. Its scope encompasses text understanding, where
algorithms dissect unstructured text for tasks like named entity recognition, sentiment
Department of CS&E, PESITM, Shivamogga. Page 8
Voice Based Email Generating System Using AI
analysis, and text classification. Furthermore, NLP delves into language modeling,
training models to grasp linguistic structures and contexts through syntax and semantic
analysis. This technology also facilitates machine translation, enabling automated
translation between languages, and supports question answering systems, chatbots, and
virtual assistants by deciphering user queries and generating coherent responses.
Additionally, NLP aids in text generation, extracting structured information from
unstructured text data, and has far-reaching applications across various industries, from
healthcare to finance, profoundly influencing the way humans interact with technology.
1.6 Django Web Framework
It is a sophisticated Web applications system that allows speedy development and
flawless, useful planning. Working with seasoned experts, it handles a big portion of the
hassle of digital marketing, allowing you to concentrate on writing your software and
having to start over. It is free and freely available. The Server - side web project's benefits
include being Super Quick, Pleasingly Reliable, Adaptable, Nearly Full, and Remarkably
Useful.
Django is a high-level web framework written in Python, designed to enable rapid
development of secure and scalable web applications. Launched in 2005, Django follows
the "batteries-included" philosophy, providing developers with a comprehensive set of
tools and libraries to streamline common web development tasks. At its core, Django
emphasizes the principle of DRY (Don't Repeat Yourself) by promoting reusable
components and modular design patterns. One of its key features is its powerful ORM
(Object-Relational Mapping) system, which simplifies database interactions and abstracts
away the complexities of SQL queries. Django also includes built-in support for user
authentication, session management, and security features such as protection against
common web vulnerabilities like SQL injection and cross-site scripting (XSS). Its robust
templating engine allows for the creation of dynamic and responsive web pages, while its
admin interface offers a convenient way to manage site content. Django's extensibility
through third-party packages and its vibrant community contribute to its popularity among
developers for building a wide range of web applications, from simple blogs to complex
enterprise systems.
Department of CS&E, PESITM, Shivamogga. Page 9
Voice Based Email Generating System Using AI
CHAPTER 2
LITERATURE SURVEY
A literature survey on "Voice Based Email Generating System Using Artificial
Intelligence" delves into existing research and technologies pertaining to this innovative
approach in communication. It explores the fusion of artificial intelligence (AI) and
natural language processing (NLP) techniques to enable systems to comprehend spoken
language and generate email content seamlessly. The survey delves into foundational
elements such as speech recognition systems, including established APIs and custom-built
models, which serve as the backbone for interpreting voice input. Additionally, it delves
into methodologies for generating email content from spoken language, encompassing
summarization techniques, sentiment analysis, and grammar checking to ensure the
coherence and relevance of the generated emails. Furthermore, the survey addresses
crucial aspects of user experience design, emphasizing the importance of intuitive
interfaces and inclusive design principles to cater to diverse user needs. Security and
privacy considerations are also explored, highlighting the necessity of robust
authentication mechanisms and data encryption to safeguard user information.
Through case studies and real-world implementations, the survey sheds light on
the practical implications and potential impact of voice-based email systems on enhancing
communication accessibility and usability. Lastly, it identifies existing challenges and
outlines future research directions to further advance this promising intersection of AI and
communication technology.
This entails an examination of user feedback and empirical studies to gauge the
usability and efficacy of voice interfaces in email communication for individuals with
diverse needs. Furthermore, the survey delves into the evolving landscape of voice
technology, considering emerging trends such as multi-modal interfaces and integration
with smart assistants to enhance the functionality and versatility of voice-based email
systems. By synthesizing insights from existing research and highlighting areas for
further exploration, the literature survey aims to provide a comprehensive understanding
of the current state and future potential of voice-based email systems empowered by
artificial intelligence.
Department of CS&E, PESITM, Shivamogga. Page 10
Voice Based Email Generating System Using AI
2.1 Survey Papers
Numerous projects have been carried out in recent years to design and analysis of
human computer interaction using AI. Here are some of the papers reviewed.
Title: The Way to Make Blind People Use the Email System: Voice Based
Email Generating System Using Artificial Intelligence [1]”
Authors Name: Gaurav kumar Rajput, Sachin Sharma ,v, Dr. Meraj
Farheen Ansari , Pawankumar Sharma.
Year: 2023.
Description:
Nowadays, Technology usage has increased, and it offers many opportunities for
present-day generations to fully embrace the Internet. Email continues to be the most
common form of communication technology in the business world. People with vision
loss find it very difficult to use this technology because their use requires visual and touch
perception. Developing computerized practice systems has opened up numerous
possibilities for the visually impaired. This system is very helpful for the blind to use
Internet applications. This project introduces the structural design of a voicemail system
that blind people can use for easy email access. The application uses speech-text and text-
speech algorithms where visually impaired users can make effective use of the
application.
Advantages:
The Python-based email interaction system provides intuitive accessibility through
voice commands, enhancing efficiency and security for users with visual impairments.
Drawbacks:
While the system offers accessibility through voice commands, users with speech
impairments or in noisy environments may face challenges, potentially limiting its
usability for some individuals with disabilities. Additional support options or alternative
input methods could enhance inclusivity and address these limitations.
Department of CS&E, PESITM, Shivamogga. Page 11
Voice Based Email Generating System Using AI
Conclusion:
While the application offers efficient email interaction for the visually impaired
through voice commands, it may present accessibility challenges for users with speech
impairments or in noisy environments. Additionally, the reliance on voice input requires
users to adapt to a new interaction paradigm, potentially affecting usability for those
accustomed to traditional interfaces.
Title: “Voice based e-mail System for Blinds [2]”
Authors Name: Pranjal Ingle, Harshada Kanade, Arti
Lanke. Year: 2020.
Description:
The aim of this paper is Internet has become one of the basic amenities for day-to-
day living. Every human being is widely accessing the knowledge and information
through internet. However, blind people face difficulties in accessing these text materials,
also in using any service provided through internet. The advancement in computer based
accessible systems has opened up many avenues for the visually impaired across the
globe in a wide way. Audio feedback based virtual environment like, the screen readers
have helped Blind people to access internet applications immensely. We describe the
Voicemail system architecture that can be used by a Blind person to access e-Mails easily
and efficiently. The contribution made by this research has enabled the Blind people to
send and receive voice based e-Mail messages in their native language with the help of a
computer
Advantages:
This system makes the disabled people feel like a normal user. They can hear the
recently received mails.
Drawbacks:
While the email system provides inclusive accessibility features, its reliance on
speech-to-text and text-to-speech functionalities may encounter accuracy issues or
compatibility challenges with diverse speech patterns or languages, potentially hindering
Department of CS&E, PESITM, Shivamogga. Page 12
Voice Based Email Generating System Using AI
effective communication for certain users.
Conclusion:
This e-mail system can be used by any user of any age group with ease of access.
It has feature of speech to text as well as text to speech with speech reader which makes
designed system to be handled by visually impaired person as well as blind person.
Title: “Voice Based Email with Security for Visually Challenged
[7]” Authors Name: Latha L, Babu B, Sowndharya S.
Year: 2020.
Description:
The aim of this paper is Communication is an important aspect of connectivity
among the people of different countries. Various communication technologies include
telephone, smart phone, internet applications such as email, what’s app, sms, etc. These
technologies are integrated with the internet. Letters were a form of communication in
olden days. They were replaced by Email (also known as electronic mail) which is a form
of dual way communication. In email, either two persons can communicate privately or
group mails can also be forwarded. But there is a problem that these internet integrated
technologies are workable only with visual perception. There is an estimation that there
are about 20 million around the world who are visually challenged. Technology
development is not only meant for normal people. The main aim of this project is to
provide an android application specially developed for the visually challenged people to
send and receive mails. To ensure privacy on sending the message, a hardware device
with human detection sensor is arranged that can be embedded into the phone using other
technology such as nanotechnology.
Advantages:
The application's inclusive design caters to both visually challenged and sighted
users, promoting accessibility in education and communication while ensuring secure
email interactions.
Department of CS&E, PESITM, Shivamogga. Page 13
Voice Based Email Generating System Using AI
Drawbacks:
The hardware-dependent privacy feature may add complexity and cost to the
application, potentially limiting its accessibility and adoption among visually challenged
users with limited resources.
Conclusion:
While the application facilitates email communication for both visually challenged
and sighted users, its reliance on hardware to ensure message privacy may introduce
complexity and cost, potentially limiting accessibility and adoption, particularly among
visually challenged users with limited resources or technical expertise.
Title: “Voice Based E-Mail System using Artificial Intelligence [6]”
Authors Name: Rijwan Khan, Pawan Kumar Sharma, Sumit Raj, Sushil
Kr. Verma, Sparsh Katiyar
Year: 2020.
Description:
One of the most prevalent forms of communication is email, serving as a vital
channel for exchanging confidential and urgent information. However, approximately 253
million visually impaired individuals worldwide encounter communication barriers. To
address this challenge, the authors propose a Voice-based Email System using AI, aiming
to enhance accessibility for visually impaired individuals and promote societal
inclusivity. Emphasizing accessibility as a pivotal feature, the system ensures ease of use
for both able-bodied and disabled individuals, aligning with the principle that true
accessibility encompasses universal usability.
Advantages:
The proposed system enables visually challenged individuals to actively engage
in digital communication, fostering inclusivity and integration into society.
Department of CS&E, PESITM, Shivamogga. Page 14
Voice Based Email Generating System Using AI
Drawbacks:
While the system promises significant benefits for visually challenged individuals,
challenges in implementation, including technical complexity and cost considerations,
may hinder its widespread adoption and exacerbate existing disparities in access to
resources and skills among this community.
Conclusion:
The proposed system's focus on fostering societal inclusion and empowering
disabled individuals, particularly the visually challenged, in the digital realm highlights
its potential to significantly enhance their quality of life and contribute to India's digital
growth. Moreover, by overcoming barriers to email communication, it could inspire
further innovation in accessibility technology, driving positive change for the visually
challenged community.
Title: “Voice based E-mail for the Visually Impaired [4]”
Authors Name: Aishwarya Belekar, Shivani Sunka, Neha Bhawar,
Sudhir Bagade
Year: 2020.
Description:
The pervasive use of technology necessitates the comprehensive utilization of
Internet resources, with email standing as a cornerstone feature. Despite the prevalence of
screen readers, visually impaired individuals encounter challenges in internet navigation.
This paper endeavors to address these obstacles by introducing voice assistance,
extending its application beyond email to encompass essential daily tools like calculators
and music players, thereby enhancing accessibility for all users.
Advantages:
The integration of speech-to-text and text-to-speech functionalities in the email
system enhances accessibility for visually impaired individuals and reduces cognitive
load, rendering it user-friendly for all users.
Department of CS&E, PESITM, Shivamogga. Page 15
Voice Based Email Generating System Using AI
Drawbacks:
The reliance on mouse clicks in papers poses accessibility challenges for visually
impaired individuals, compounded by limitations in speech recognition systems primarily
tailored for English, prompting the proposal of a system catering to the needs of the
visually impaired. It was found that participants experienced great challenges in directing
their attention to the menus and buttons, understanding the meaning of icons, and
interacting with these menu components.
Conclusion:
This paper proposes a Voice-based Email system designed to aid visually
impaired individuals, enabling independent and efficient access to email services. By
eliminating the need for keyboard shortcuts, the system empowers users to read and send
emails using voice commands, facilitated by a speech recognition application, thereby
catering to the needs of both visually impaired and other disadvantaged users.
Title: “V-Mail (Voice Based E-Mail Application) [5]”
Authors Name: Naziya Pathan, Nikita Bhoyar, Ushma Lakra,
Dileshwari Lilhare.
Year: 2019.
Description:
Our routine is initiated by the Internet. It is the First thing in the morning we do
see our Notification and EMails. The internet has made human life so much easier, now
the biggest and toughest tasks are done in minutes. No matter it is a simple pizza order,
shopping or money transfer, and communicates with the help of emails it is so much
easier by the use of Internet in life. But there is a special criterion for humans to access
the Internet and the criteria is you must be able to see. But visually challenge people
cannot able to access such types of communication and technologies on their own. Vmail
helps blind people to access e-mail.
Department of CS&E, PESITM, Shivamogga. Page 16
Voice Based Email Generating System Using AI
Advantages:
The proposed voice-based email system for visually impaired individuals
enhances accessibility, reduces cognitive load, and fosters empowerment through
intuitive voice commands, promoting universal inclusivity and productivity.
Drawbacks:
One potential disadvantage of the proposed voice-based email system for visually
impaired individuals could be the dependency on reliable internet connectivity and voice
recognition accuracy, which might pose challenges in environments with limited
connectivity or for users with speech impediments.
Conclusion:
The planned system aims to streamline email access for visually impaired
individuals by eliminating the need for keyboards through screen readers and voice
commands, potentially overcoming previous accessibility challenges and extending its
utility to other tasks, yet it may still require reliance on consistent internet connectivity
and accurate voice recognition.
2.2 Summary of the literature survey
The literature survey provides an overview of the research and development
related to Voice Based Email Generating System Using Artificial Intelligence projects.
The aim is to explore the current state of the art, identify key technologies,
methodologies, and challenges, and highlight recent advancements in the field. The
survey covers a wide range of academic and industry publications, including journal
articles, conference papers, and technical reports. The findings of this survey will serve
as a valuable resource for researchers, engineers, and policymakers involved in the
design, implementation, and evaluation of Voice Based Email Generating System Using
Artificial Intelligence projects.
Department of CS&E, PESITM, Shivamogga. Page 17
Voice Based Email Generating System Using AI
2.3 Problem Description
Problem statement: “Lip synthesis may potentially serve as an efficient informational
method for addressing devices for the blind. We started working on our project to
make it possible for people with disabilities to reach their destinations without
difficulty using a significant associations route. Various mechanical modifications
have indeed been put in place for those who are physically weaker so that they can
utilize and gain from mobile phone improvements. Applications that help blind have
sent and read texts will just be created using a.i., taking this into consideration that it
is a crucial idea. In addition to maintaining a selection of words to comprehend it, we
want to nurture a computer forensics toolkit for the disabled via our research.”
Department of CS&E, PESITM, Shivamogga. Page 18
Voice Based Email Generating System Using AI
CHAPTER 3
APPROACHES AND METHODS
Developing a voice-based email generating system using artificial intelligence
(AI) involves several key steps. Initially, speech recognition technologies are employed to
accurately transcribe spoken words into text. Natural Language Processing (NLP)
techniques are then utilized to understand the transcribed text, including tasks such as
tokenization, part-of-speech tagging, and syntactic parsing. Intent recognition
mechanisms are crucial for identifying the user's intention, whether it's composing a new
email, replying, or forwarding. The heart of the system lies in email content generation,
where AI models generate relevant and contextually appropriate email content.
Personalization further enhances the user experience by tailoring emails to individual
preferences and historical interactions. Grammar and style checking ensure the final email
adheres to linguistic standards. Integration with email APIs enables seamless sending,
with security measures in place to protect user data. A feedback mechanism collects user
input for continuous improvement, guiding future iterations of the system. Through
iteration and adaptation, the system evolves to meet user needs and advances in AI
technology.
3.1 Proposed Method
The addressing will be sent by e - mail address, making it readily available to
those with poor eyesight and beneficial to the neighbourhood. A level implementation is
allegedly inadequate and entirely open, independent of how it is used in visual acuity.
As a competitor of the current structure, they can tend to focus on the usefulness
of typical customers as the package's merits, the concern for acceptability by a broad
variety of people, the typical persons, and the legally blind.
For the purpose of using the services, this foundation obligates the customer to do
some action, and if the terms of receiving a number of services, they must perform this
task. The customer will first sign up for the program using the enrollment form. The buyer
must be completely completed using audio signals and the required forms before it is
evaluated on the website and stored as the customer speaks
Department of CS&E, PESITM, Shivamogga. Page 19
Voice Based Email Generating System Using AI
After it user logs in, the program will ask for their identification and encryption
algorithm, and converts their voice become words. The user is then verified by looking up
their credentials in the data base. You may entice the consumers in the specific segments
to document your prospecting activity and communications sent after a successful.
Figure 3.1: Flowchart of user login and accessing
Department of CS&E, PESITM, Shivamogga. Page 20
Voice Based Email Generating System Using AI
The Figure 3.1 is the suggested method will be accessible to those with poor vision
via email and will benefit the neighbourhood. The authors claimed that anyone who is
currently a part of this method will find it simple. It is said that a detailed application is
ineffective and completely accessible wherever it is used in human vision.
As a critic of he present system, they would rather have features that made the
system easier to use for traditional users, with special attention paid to the needs of the
visually impaired and traditional users in Africa.
For the use of the services, this method requires a certain action from the user, and
if the user has access to a number of services, they will also need to take this action. The
user will first fill out the enrollment form to sign up for the application. The user must be
filled out completely using voice commands and all necessary fields before being scanned
in from the website; as soon as the user starts speaking, an automatic recording will begin.
The system will prompt the user for a user name and password after the user logs
in, translating the audio into text, and the user will then be verified by verifying the
database credentials. After a successful test, you can ask the users in the various areas to
record your sent and received messages.
Department of CS&E, PESITM, Shivamogga. Page 21
Voice Based Email Generating System Using AI
CHAPTER 4
APPLICATIONS
Here are some potential applications of a voice-based email generating system using
artificial intelligence for blind individuals:
Independent Communication: Blind individuals can use the system to send and
receive emails independently without relying on assistance from others.
Work and Business Communication: They can use the system for professional
communication, such as sending work-related emails, responding to client
inquiries, or scheduling meetings.
Personal Correspondence: The system enables blind users to stay connected with
friends and family through email, allowing them to send personal messages, share
photos, and stay updated on important events.
Accessibility in Education: Blind students can use the system to communicate
with professors, submit assignments, and participate in online discussions,
enhancing their accessibility and inclusion in educational settings.
Access to Information: By accessing their email accounts through the system,
blind individuals can stay informed about news, events, and updates from
organizations or newsletters they subscribe to.
Integration with Assistive Technologies: The system can be integrated with other
assistive technologies, such as screen readers or braille displays, to provide a
seamless and accessible email experience for blind users.
Accessibility in Online Shopping: Blind users can utilize the system to
communicate with customer service representatives, track orders, and receive order
confirmations or shipping updates via email.
Healthcare Communication: They can schedule appointments, communicate with
healthcare providers, and receive medical reminders or test results through email
using the system.
Access to Government Services: Blind individuals can interact with government
agencies, submit forms or applications, and receive notifications or updates about
their benefits or legal documents via email.
Department of CS&E, PESITM, Shivamogga. Page 22
Voice Based Email Generating System Using AI
CHAPTER 5
RESULTS
Developing a voice-based email generating system using artificial intelligence
(AI) involves integrating various technologies and methodologies. At its core, the system
employs speech recognition to transcribe spoken language into text. This transcription is
then processed using natural language processing (NLP) techniques to extract meaning
and intent from the user's input.
The system must accurately recognize the user's commands, such as composing a
new email, replying to an existing one, or forwarding a message. Once the user's intention
is understood, AI algorithms generate the email content based on the provided input and
contextual information. This could involve accessing relevant data sources, including
previous email conversations or external information.
Figure 5.1: Login Page
Figure 5.1 shows the live server login page of the voice- based email system for
the blind person. The consumer needs a Youtube confirmed Hotmail login in order to log
in to the messaging system, and they should use the mouse to select any location on the
display to activate the internet state's quality issue.
By clicking anywhere on the screen, the voice assistant gets activated and ask user
to speak their existing email id and password.
Department of CS&E, PESITM, Shivamogga. Page 23
Voice Based Email Generating System Using AI
Figure 5.2: Main Menu Page
Figure 5.2 Shows once account is verified the user is then directed to the menu
page.
Figure 5.3: Compose Mail Page
Figure 5.3 Shows by again clicking anywhere on the screen the voice assistant
again gets activated and ask user to speak for his next move. If user speak for ‘compose’
then he is directed to the compose mail page.
Department of CS&E, PESITM, Shivamogga. Page 24
Voice Based Email Generating System Using AI
CHAPTER 6
CONCLUSION
Our software's purpose is to assist those who are blind or visually impaired by
using voice communication to send texts. Anyone of all ages who is ready to get involved
may utilise our mobile messenger app. It makes use of several features, such as the
conversion of spoken language into written text and vice versa. Blind persons who use a
device may be able to communicate with it by speaking into it. This device will then
translate what users say in and out of letters and use a computer-produced voice to direct
them as to where they are in the document and what they must do next. Inside this way,
this device might be quite useful for those who are physically weak. By using voice-based
emails, it may be made available to everyone who is visually impaired. The technology is
straightforward and simple enough blind individuals to use without requiring a laptop.
Department of CS&E, PESITM, Shivamogga. Page 25
Voice Based Email Generating System Using AI
REFERENCES
[1] Gaurav kumar Rajput, Sachin Sharma ,v, Dr. Meraj Farheen Ansari , Pawan kumar
Sharma, Dr. Surendra Kumar Shukla. The Way to Make Blind People Use the
Email System: Voice Based Email Generating System Using Artificial
Intelligence.
[2] Khan, R., Sharma, P. K., Raj, S., Verma, S. K., & Katiyar, S. (2020). Voice Based
E-Mail System using Artificial Intelligence. International Journal of Engineering
and Advanced Technology (IJEAT).
[3] Ingle, P., Kanade, H., & Lanke, A. (2016). Voice based e-mail System for Blinds.
International Journal of Research Studies in Computer Science and Engineering
(IJRSCSE).
[4] Latha, L., Babu, B., & Sowndharya, S. (2020). VOICE BASED EMAIL WITH
SECURITY FOR VISUALLY CHALLENGED.
[5] Kulkarni, O., Alhat, A., Tejankar, N., & Patil, M. (2019). VOICE BASED E-
MAIL SYSTEM FOR BLIND PEOPLE. Open Access International Journal of
Science and Engineering.
[6] Rijwan Khan, Pawan Kumar Sharma, Sumit Raj, Sushil Kr. Verma, Sparsh Katiyar.
"Voice Based E- Mail System using Artificial Intelligence". International Journal of
Engineering and Advanced Technology (IJEAT) ISSN: 2249– 8958, Volume-9, Issue-
3, February, 2020.
[7] Nilesh, J., Alai, P., Swapnil, C., & Bendre, M. (2014). Voice based system in
desktop and mobile devices for blind people. International Journal of Emerging
Technology and Advanced Engineering (IJETAE).
[8] Jagtap Nilesh, Pawan Alai, Chavhan Swapnil, and Bendre M.R."Voice Based
System in Desktop and Mobile Devices for Blind People". In International Journal
of Emerging Technology and Advanced Engineering (IJETAE).
Department of CS&E, PESITM, Shivamogga. Page 26
