The problem at hand is to address the challenges faced by visually impaired individuals when navigating their surroundings independently. There is a pressing need for an innovative AI-driven system that leverages advanced optical technologies and artificial intelligence algorithms to enable the blind to perceive and understand their surroundings in real-time as they walk.
The visually impaired community faces numerous challenges in their daily lives, ranging from accessing institutions to participating in e-commerce activities independently. Additionally, the global issue of climate change poses a growing threat to all, including the blind population.

Gemini AI Hackathon

Bridging Accessibility and Awareness for the Blind using Cutting edge technologies to enhance independent mobility.

Shield Team

Gemini AI

Google

YOLOv6

YOLO

Llama 2

TruLens

TruEra

GPT-4 Vision

OpenAI

BlindEye

YouTurbio allows you to analyze videos with Video Citations, and evaluate different tasks and metrics:

We use LlamaIndex + GPT-4 for chunks classification leveraging function calling
MongoDB as vector Database
Cohere as QA LLM 

This tool helps you understanding the mood of a video, and directly linking to the information source.

You can use any topic categories you want, and then ask questions about the video content, or the classified topics.

After you finish, you can take a look at the TruLens evaluation, to make adjustments and improve the quality of your system!.

TruLens Leaderboard:

QA: Evaluates Cohere's performance answering questions
Classify: Evaluates GPT's performance classifying video transcript chunks

YouTurbio allows you to analyze videos with Video Citations, and evaluate different tasks and metrics

Kibernum USA

One of the best AI labs in the world. By using the Cohere platform we are able to use one of the best and most powerful language models for our own! There are many tutorials, including our own one, that show how to use Cohere.

Cohere

Cohere is a Canadian startup that provides natural language processing models that help companies improve human-machine interactions

OpenAI is an AI research and deployment company

Empower your AI projects with TruLens, a suite of tools for developing and evaluating large language models.

Enabling trust in AI models with TruEra's quality solutions.

LlamaIndex connects data sources to large language models, providing tools to ingest, index, and query data for building powerful AI apps augmented by your knowledge.

LlamaIndex

YouTurbio - Video analysis and Evaluation

* Business Value and Market Impact
- AI Story Maker opens new opportunities in the growing digital book publishing industry, estimated at $22 billion.
- Authors often struggle with the time-consuming process of Amazon KDP niche research, content creation, and optimization for Amazon KDP. 
- "AI Story Maker" addresses the growing demand for automated solutions in content creation, particularly in the competitive book publishing industry. By reducing the time and expertise required to research, write, and optimize books for Amazon KDP, this platform enables authors and publishers to focus more on creative expression and marketing, rather than the nuances of niche research and initial content drafting. 

* Technology Stack and Implementation
Our MVP for AI Story Maker harnesses the full capabilities of our sponsor's technology stack, ensuring an advanced, reliable, and insightful platform. Here’s how we’re utilizing MongoDB, TruLens, and Llama-Index in developing our project:

- MongoDB Atlas: Hosting and querying a dataset of Amazon book keywords and KDP guidelines for optimized content creation.
- TruLens: Evaluating the dataset quality on groundedness and relevance, ensuring high-standard content output.
- Llama-Index: Enabling real-time data querying for dynamic content generation based on user inputs.

* How it Works: From Idea to Finished Book
- Users start by entering parameters like story idea, size, character details, and genre into our user-friendly interface.
- AI Story Maker generates a detailed book proposal, including optimized title, description, and keywords, as well as a chapter outline.
- After reviewing the proposal, if the user approves, our AI then proceeds to write full chapters. These chapters are formatted in markdown and can be downloaded directly for review or publishing.

AI-driven platform for Amazon KDP that automates niche research and book content creation with human co-creation.

Krebit Team

GPT-3.5 stands for Generative Pre-trained Transformer and is fine-tuned version of GPT-3. It is an autoregressive language model that uses deep learning to produce human-like text. There are many models in GPT-3.5 family as gpt-3.5-turbo or gpt-3.5-turbo.

GPT-3.5

A powerful and flexible document-based NoSQL database that has gained popularity among developers.

MongoDB

AI Story Maker - Book Creator Assistant

💼In the realm of traditional diagnostics, several challenges loom large. Patients often struggle to assess symptoms accurately, leading to worsened illnesses. Healthcare accessibility remains limited, with barriers such as time and finances preventing many from seeking necessary care. The imbalanced workforce places undue pressure on doctors, potentially compromising patient care.🚀Enter DeepSymp+: An evidence-based 
 diagnostic AI poised to revolutionize healthcare. This platform offers precise symptom evaluation, benefiting both patients and providers, and fostering a more efficient healthcare system.🛠️ We constructed DeepSymp+ using a combination of technologies, including Streamlit, LlamaIndex, MongoDB, ChatGPT, and Trulens. Leveraging OpenAI's LLM, we equipped our AI with the ability to provide medical advice based on a vast database of knowledge. Each component underwent rigorous testing and optimization using Trulens evaluation functions.🏋️‍♂️Initially, our AI faced three primary challenges: Accuracy, Context Faithfulness, and Cost Efficiency. Through diligent testing and optimization, such as enriching our medical knowledge database and optimizing our pipeline with Trulens, we successfully overcame these hurdles.🏅We take pride in developing a diagnostic AI that substantiates its evaluations with medical evidence from textbooks and research papers, contributing to healthcare improvement.🔮Our journey doesn't end here. We'll continue innovating and iterating upon DeepSymp+. Our plans include building a web application, ensuring model accuracy and safety through thorough testing, and ultimately launching the project. We remain committed to refining and enhancing our groundbreaking diagnostic AI.

Introducing DeepSymp+: A symptom assessment AI revolutionizing healthcare. DeepSymp+ analyzes symptoms, offering precise preliminary diagnoses and streamlining medical processes.

Average Fun Enjoyer

ChatGPT is a large language model trained by OpenAI to generate human-like text in a conversational style. It is a variant of the GPT-3 model, which was specifically designed to be used to generate text in response to user input. The ChatGPT model has been trained on a vast amount of text data, including conversations and other types of human-generated text, which allows it to generate text that is similar in style and content to human conversation. ChatGPT can be used to generate responses to questions, code, make suggestions, or provide information in a conversational manner, and it is able to do so in a way that is often indistinguishable from human-generated text. The initial model has been trained using Reinforcement Learning from Human Feedback (RLHF), using methods similar to InstructGPT, but with slight differences in the data collection setup. The model is trained using supervised fine-tuning, where human AI trainers provided conversations in which they played both sides—the user and an AI assistant. The trainers would have had access to model-written suggestions to help them compose their responses.

ChatGPT

DEEPSYMP - Evidence-Based Diagnostic AI

Better Health AI attempts to improve the quality of care by offering different payment models for different health services that ties payment to quality of health outcome while trying to reducing the cost. There have been many care delivery and payment models attempted at state and federal level targeting various diseases, prevention, safety, infection, and other areas of health. The result has been mixed and some are undetermined. There have been many studies and reports published. 

This project reviews different studies, published articles and reports. It reviews user's scenario of health services and recommends suitable models and implementation strategy with supported facts and figures. The user can work with the application interactively to evaluate different models with success matrix and references for easier decision making. Reducing cost while trying to improve the quality of care may work counter to each other. This project helps to strike the balance and provide examples and approach to achieve this simultaneously.

Better Health AI is a Recommender Service in the health care domain that explores and recommends Alternative Payment Model (APM) and Value Based Care (VBC) models.

BetterHealthAI

Simplify AI integration with OpenAI's Assistants API, empowering developers to create AI assistants effortlessly.

Assistants API

Fronto is a revolutionary AI-powered website builder designed to simplify the website creation process for individuals and businesses. With innovative features like "ScreenShot to code" and "Image Link to code," Fronto eliminates the need for manual coding and design, allowing users to create professional websites quickly and easily. Leveraging advanced AI assistants like GPT-4 and GPT-4 Vision and another models, Fronto automates the entire process, from design inspiration to functional code generation. Trulens provides feedback and evaluates the performance of our OpenAI assistants, ensuring the effectiveness and reliability of Fronto's AI-powered features. Whether you're a novice or an experienced developer, Fronto empowers you to bring your website ideas to life effortlessly.

Fronto: AI-powered website builder with features like 'ScreenShot to code' & 'Image Link to code', enabling quick, easy website creation.

AEye

Explore the groundbreaking GPT-4 Vision, empowering AI to understand and process images, revolutionizing capabilities across diverse domains.

Explore image generation capabilities with the cutting-edge DALL·E API.

DALL·E Image Generation API

Clarifai's groundbreaking technology aims to bring The World’s AI to developers and data scientists around the globe.

Clarifai

GPT-4 is OpenAI's 4th generation Generative Pre-trained Transformer. It is a multimodal large language model that uses deep learning to produce human-like text, accepting image and text inputs. GPT-4 is OpenAI's most advanced system, producing safer and more useful responses