Sign Language System Requirements Document

Introduction

The "Sign Language" project aims to facilitate communication between individuals with hearing and speaking impairments and those who do not understand sign language. This system will enable bidirectional communication by converting gestures into text and text into sign language, using datasets from Kaggle focused on Indian Sign Language.

System Overview

The system will provide a seamless communication platform where individuals with hearing and speaking impairments can use gestures to communicate with those who do not understand sign language. The project will analyze gestures and convert them into text for non-sign language users. Conversely, it will also convert text or spoken language from non-sign language users into sign language, displayed visually to the impaired individuals. This dual functionality not only aids communication but also serves as an educational tool for learning sign language.

Functional Requirements as Story Points

• As a hearing/speaking impaired user, I should be able to perform gestures that the system can analyze and convert into text for non-sign language users.
• As a non-sign language user, I should be able to input text or speech that the system can convert into sign language gestures for the impaired user.
• As a user, I should be able to learn sign language through the text-to-sign language conversion feature.
• As a developer, I should be able to use nvm for managing Node.js versions instead of npm.

User Personas

• Hearing/Speaking Impaired User: Individuals who use sign language to communicate.
• Non-Sign Language User: Individuals who do not understand sign language but wish to communicate with impaired users.
• Developer: Individuals responsible for maintaining and updating the system.

Core User Flows

• Hearing/Speaking Impaired User performs gesture -> System analyzes gesture -> Converts to text -> Non-Sign Language User reads text.
• Non-Sign Language User inputs text/speech -> System converts to sign language -> Displays gestures -> Hearing/Speaking Impaired User understands message.

Visuals Colors and Theme

• primary: #1A73E8 (Deep Blue)
• primary_light: #4D90FE (Light Blue)
• secondary: #FF7043 (Coral)
• accent: #FFCA28 (Amber)
• highlight: #FFA726 (Orange)
• bg: #F5F5F5 (Light Gray)
• surface: rgba(255, 255, 255, 0.8) (White)
• text: #212121 (Dark Gray)
• text_muted: #757575 (Medium Gray)
• border: rgba(0, 0, 0, 0.1) (Light Black)

Signature Design Concept

The homepage will feature an interactive "Gesture Galaxy" where each star represents a different sign language gesture. Users can click on a star to see the gesture animated, with accompanying text explaining its meaning. As users interact with the galaxy, the stars will rearrange to form constellations representing common phrases. This dynamic and educational experience will be built using @react-three/fiber for 3D interaction and gsap for smooth animations.

Landing Hero Motion Brief

The landing hero will depict a continuous loop of a hand performing a series of sign language gestures. Each gesture will transform into a corresponding text message, illustrating the conversion process. This animation will be created using motion/react to ensure smooth transitions and interactions. The scene will be responsive, adapting to different screen sizes while maintaining clarity and engagement.

Interaction Model & Motion Direction

• Intended Interaction Model: Parallax
• The landing page will feature layered depth via scroll, with decorative elements translating at different speeds to create a visually rich first impression. Interactive elements will have hover transitions and spring physics for a polished user experience.

Non-Functional Requirements

• The system should support real-time gesture recognition and text conversion.
• The system should be scalable to accommodate a growing number of users.
• The system should ensure data privacy and security for all users.

Tech Stack

• Frontend: React for Web
• Backend: Python, FastAPI
• Database: MySQL or MariaDB
• AI Models: Google Nano banana for image generation
• AI Tools: Litellm for LLM Routing, Langchain
• Local Orchestration: Docker, docker-compose
• Server-side Orchestration: Kubernetes

Assumptions and Constraints

• The system will use datasets from Kaggle focused on Indian Sign Language.
• The system will utilize nvm for Node.js version management.
• The system will require internet connectivity for real-time processing.

Glossary

• Gesture Recognition: The process of interpreting human gestures via mathematical algorithms.
• Sign Language: A language which uses manual communication and body language to convey meaning.
• nvm: Node Version Manager, a tool for managing multiple Node.js versions.