CS8611-Mini Project - PPT Template-4 (1).ppt

CS8611- MINI PROJECT VIVA VOCE – APRIL/MAY-2022 EXAMINATIONS
1
SIGN LANGUAGE TRANSLATOR USING TRANSFER LEARNING
BATCH MEMBERS
1.JANARTHANAN P (513420104030)
2.VIGNESH S (513420104312)
3.SUSEENDHERAN M (513420104702)
GUIDED BY – T.KALA

With the increase of innovations and technology, life has become significantly easy for humans. The sudden
surge of growth in tech has left many overjoyed and overwhelmed because of the good fruits it bears. It has
paved way for the poor to b88ecome rich, the sick to become strong, the disabled to experience the life of an
abled. People with speech/hearing impairment have always found it difficult to communicate and mingle but
with technology, that barrier has also been destroyed. They can now communicate without any difficulty and find
themselves in a public setting, communicating confidently.
CS8611- MINI PROJECT VIVA VOCE – APRIL/MAY-2022 EXAMINATIONS 2

According to the World Health Organization (WHO), 466 million people across the world have disabling
hearing loss (over 5% of the world's population), of whom 34 million are children. There are only about 250
certified sign language interpreters in India for a deaf population of around 7 million. With these significant
statistics, the need for developing a tool for smooth flow of communication between abled and people with
speech/hearing impairment is very high. Our application promises to secure a two way conversation, as it
deploys machine learning and deep learning models to convert sign language to speech/text. The opposite
receiver can either speak or text his response, which will then be visible to the disabled person in the form of
text. The client can make use of the tutorials and learn the basic functioning of the application and ASL. This
system eliminates the need of an interpreter and the traditional methods of pen and paper can also be discarded.
This application ensures the automation of communication and thereby provides a solution to the hurdles faced
by hearing/speech impaired people.

1) Eliminate the need of an interpreter.
2) Ease the communication flow for hearing/speech impaired people through our model predictions and text to
speech system.
3) Provide high quality video tutorials for most commonly used phrases and English sign alphabets (ASL).
4) Provide a portal for practicing the learned alphabets.
5) Ability to create new signs for any text or sentence in the browser (client side).

 Sign Language Translator enables the hearing impaired user to communicate efficiently in sign
language, and the application will translate the same into text/speech. The user has to train the model,
by recording the sign language gestures and then label the gesture. The user can then use the saved and
recorded gestures while speaking to other people.

The traditional methods of communicating with the deaf and mute are really not convenient in many aspects. The
alternatives that are available to break this barrier have definite flaws. 2 An interpreter is not always available
and this method is not cost efficient either. The pen and paper method is highly unprofessional and also time
consuming. Texting and messaging are fine to a certain extent but still does not tackle the bigger problem at
hand. This has created a grave need to develop a solution to destroy the barricade of communication effectively.

 1) The user has to himself provide the dataset by making the sign language gestures and then label them.
When the user records a large set of gestures.
 2) For practicing skills it is necessary for the user to have a white background with no other object in the
frame of the camera apart from the hand

 Recognition and Tracking of Sign Language: Stoll et al. (2018) presented a comprehensive review of
computer vision techniques for sign language recognition and tracking. The study discusses various approaches,
such as depth sensors, skeletal tracking, and convolutional neural networks (CNNs), for accurate hand and
gesture recognition.
 Sign Language Datasets and Corpora: Athitsos et al. (2008) proposed a methodology for building large-scale
sign language datasets. The study highlights the importance of data collection, annotation, and representation
for training and evaluating sign language recognition systems.
 Gesture Recognition and Machine Learning: Starner et al. (1998) introduced the concept of "Gloves-on" sign
language recognition, utilizing data gloves equipped with sensors to capture hand movements. The study
emphasizes the use of machine learning algorithms for real-time recognition of sign language gestures
8

 Sign Language Synthesis: Koller et al. (2018) focused on the synthesis of sign language animations from
text or speech input. The research explores rule-based and statistical machine translation methods to
generate accurate and expressive sign language animations.
 Sign Language Translation on Mobile Devices: Huang et al. (2020) developed a sign language translation
system specifically for mobile devices. The research focuses on optimizing the computational efficiency of
the translation model to enable real-time translation on resourceconstrained devices

 To break the above mentioned barricade, the application promises to secure a perfect communication.
The hearing impaired user will be given access to fullfledged tutorials to guide him/her to use the
application on the website. The tutorials page for sign language consists of 70+ most common used
phrases like "Open the door " , all English alphabet in sign language and some signs for objects and
place names like Mumbai

SYSTEM REQUIREMENTS
 HTML
 CSS
 JAVA SCRIPT
 Web Browser with Internet Connection.
 Web Camera 4.2
HARDWARE REQUIREMENTS
 Processor : intel 5
 Motherboard : intel 915gvsr chipset board
 Ram : 4 GB DDR 2 ram
 Hard disk drive : GB

This project was undertaken to solve the underlying issue faced by hearing and speech impaired people. They
often don’t even stand a chance in the competitive global arena because of communication hurdles. This
project, however, helps in eradicating the social stigma of them not able to participate in many domains and
successfully gives them confidence to stand upright in any field they want. The application provides the
necessary platform to communicate with much ease and gives them the ability to interact without any external
help. The need of an interpreter is eradicated and the smooth flowing of a conversation is well developed.

The model and text to speech can be embedded into a video calling system. Thereby allowing the user to show
the gestures and the receiver on the call will receive the message in the form of text or speech. While the
receiver responds, the message will be relayed to the hearing/speech impaired user via text (subtitles).

1. Recognition and Tracking of Sign Language: a. Stoll et al. (2018) presented a comprehensive review
of computer vision techniques for sign language recognition and tracking.
2. Sign Language Datasets and Corpora: a. Athitsos et al. (2008) proposed a methodology for building
large-scale sign language datasets
3. Gesture Recognition and Machine Learning: a. Starner et al. (1998) introduced the concept of
"Gloves-on" sign language recognition, utilizing data gloves equipped with sensors to capture hand
movements.
4. Sign Language Synthesis: a. Koller et al. (2018) focused on the synthesis of sign language
animations from text or speech input.
5. Neural Machine Translation for Sign Language: a. Camgöz et al. (2018) proposed a neural machine
translation (NMT) approach to translate spoken and written languages into sign language. 17

CS8611-Mini Project - PPT Template-4 (1).ppt

More Related Content

Similar to CS8611-Mini Project - PPT Template-4 (1).ppt (20)

Recently uploaded (20)

CS8611-Mini Project - PPT Template-4 (1).ppt