EyeTalk

Optical Character Recognition(OCR)

We have integrated optical character recognition (OCR) technology into our system to support visually impaired individuals in accessing textual content from images. This advancement enables users to hear the text embedded within images, thereby enhancing accessibility and fostering greater independence in navigating printed materials. By harnessing OCR capabilities, our solution empowers visually impaired individuals to convert visual information into audible text, facilitating easier comprehension and engagement with written content.

Currency Recognition

We've developed a currency detection model tailored to assist visually impaired individuals in swiftly identifying Indian currency denominations. Using cutting-edge image recognition technology, our model offers real-time detection capabilities, instantly recognizing various currency notes. We have used pyttsx3, a text-to-speech library, enabling the detected currency to be relayed audibly to users. This not only enhances accessibility but also promotes independence and confidence in managing financial transactions for visually impaired individuals.

Natural Language Processing(NLP) using Spacy

We have leveraged the power of spaCy, a natural language processing (NLP) library, to enhance accessibility for visually impaired individuals. By integrating spaCy into our system, we have enabled intuitive interaction through voice commands, allowing users to effortlessly communicate and engage with the system's functionalities.SpaCy's robust NLP capabilities enable us to interpret and understand spoken commands, empowering visually impaired individuals to perform a variety of tasks efficiently and independently.

Optimisation of Code

This document outlines the refactoring process for the SixthSense project, focusing on the conversion of the initial script-based implementation into a modular, object-oriented design. The refactor aimed to improve code organization, readability, and maintainability by encapsulating related functionalities into classes and separating concerns into different modules.

Gemini And Gemma

Face Distance Calculation

We've developed a Face Distance System. By seamlessly integrating face detection technology into our solution, we've created a transformative experience that empowers individuals with visual challenges. This system intelligently activates when someone approaches a blind individual within a predefined distance, providing real-time auditory feedback about the identity of the approaching person. Through this groundbreaking approach, we're not only fostering independence but also promoting inclusivity, enabling visually impaired individuals to navigate social interactions with confidence and ease in a more accessible world.

Emotion Detection balance

In this system, we integrate emotion detection technology with facial recognition to assist visually impaired individuals in perceiving the emotions of people around them. When a known person stands in front of a blind individual, our system utilizes a camera feed to recognize their facial expressions. This recognition process is initiated only when the system identifies a familiar face. Once a face is detected and identified, our technology analyzes the facial expression using advanced emotion detection algorithms. Subsequently, the system converts this emotional insight into spoken words through a speech synthesis engine, enabling the blind individual to understand the emotional state of the person in front of them. Through this innovative integration of technology, we aim to enhance the social interactions and situational awareness of visually impaired individuals, fostering a more inclusive and connected environment.

Gesture Recognition

Our project introduces a groundbreaking gesture recognition system designed to empower individuals with visual impairments by providing crucial information about their surroundings. This innovative system enables users to interact with their environment through hand gestures, offering real-time object detection feedback. For instance, when a visually impaired individual closes their hand, the system initiates object detection, providing auditory or tactile cues about nearby objects. Conversely, when the hand is opened, the system ceases object detection, ensuring privacy and minimizing distractions. By leveraging gesture recognition technology, our system aims to enhance the independence and safety of visually impaired individuals, enabling them to navigate and interact with their surroundings more confidently and efficiently.

Face Recognition

We've engineered a face recognition system designed to enhance their social interactions and overall sense of autonomy. This innovative technology represents a significant leap forward in accessibility, enabling blind individuals to discern and recognize the faces of those in their immediate vicinity. By harnessing the power of artificial intelligence and computer vision, our system provides invaluable assistance to individuals who rely on auditory cues to navigate the world around them.

Object Detection

We've harnessed the potential of object detection technology to revolutionize accessibility for the visually impaired community. By integrating state-of-the-art object detection algorithms into our system, we've created a seamless experience where users can effortlessly perceive their surroundings through auditory feedback. Object detection enables our system to identify and describe nearby objects in real-time, providing users with vital information about their environment. This innovative approach not only enhances independence but also promotes inclusivity by empowering visually impaired individuals to navigate the world with confidence and ease. Through cutting-edge technology and a commitment to accessibility, we're bridging the gap between sighted and non-sighted individuals, creating a more inclusive society for all.

Speech To Text

We have built a speech-to-text system tailored for individuals with visual impairments. Our system seamlessly integrates with Google's powerful speech recognition technology, allowing users to convert spoken words into text format. This innovation aims to enhance accessibility and independence for visually impaired individuals in their daily lives. With a focus on online connectivity, our system ensures access to Google's speech-to-text service whenever network connectivity is available. However, we understand that internet connectivity may not always be reliable. In such cases, our system seamlessly switches to a whisper tiny offline model, ensuring uninterrupted assistance for visually impaired individuals, regardless of internet availability.

Voice-Enabled Calendar Management

We have integrated Google Calendar API in our Sixth Sense to provide vital resources for visually impaired individuals, enabling them to manage schedules and tasks effortlessly through voice commands. With the ability to create events on specific days and inquire about their schedule for any given day, users gain instant access to crucial information, enhancing their organization and time management skills. Additionally, the feature allowing users to make notes by simply dictating them aloud offers a convenient way to capture important thoughts and reminders.

Distance Warning System

We have implemented a distance warning system designed to assist visually impaired individuals in navigating their surroundings safely. Leveraging the advanced depth estimation capabilities of MidasV2 with a single camera setup, our system accurately calculates distances between the user and surrounding obstacles in real-time. With prompt audible alerts generated upon detecting obstacles within a customizable threshold distance, our system ensures timely warnings to prevent potential collisions. Additionally, incorporating an offline model guarantees uninterrupted assistance even in areas with limited internet connectivity, enhancing accessibility and independence for visually impaired individuals.

Text To Speech

We've developed a robust text-to-speech solution utilizing pyttsx3, tailored to aid visually impaired individuals in accessing device outputs audibly. This innovation empowers users to hear device-generated text, enhancing accessibility and promoting independence in engaging with digital content.

Features