EduInnovators

Harsh Singh

Harsh Singh

Bengaluru, Karnataka

A web app profile is created for each teacher. After entering the profile, the teacher uploads ppt, title of the slides is fetched using the nltk package in Python, using the chatGPT API explanations are fetched and stored. ...learn more

Project status: Under Development

Artificial Intelligence

Intel Technologies
oneAPI

Docs/PDFs [1]Code Samples [1]

Overview / Usage

Welcome to Smart EduInnovators on GitHub! Our project features an AI Trained Bot for advanced speech interaction, efficient information retrieval, and personalized voice recognition. Join us in revolutionizing education through cutting-edge technology.

Problem Statement

Teachers always have to attend class. On some odd day, if they are absent, the university has to appoint their replacement for that duration. In this process workload on teachers is increasing daily, and universities also have to pay extra money for the same. So here we are with the machine learning model, which is encrypted with the teacher’s voice, and the teacher can upload their content here. So that it will explain the whole content to the student for a flexible duration. The text conversion module offers three functionalities as

Personalization: allow users to create and save personalized voice profiles. Batch processing: implement the ability to process many texts in one go. Online/offline mode: offer both online and offline modes.

The task of the teacher here is to upload content here and they just have to mention their time duration here, so it will explain the whole content for that duration.

Methodology / Approach

A web app profile is created. It contains individual profiles for each teacher. After entering the profile, the teacher has the option to upload the prepared presentation for the upcoming lecture on which they have planned to take a leave. Once the ppt is uploaded, the tile of the slides are fetched using the nltk package in Python. Then using the chatGPT API, explanations are fetched for those titles and appended into a .docx file. This document generates the final audio output using the text-to-speech conversion model.

Technologies Used

  1. Machine Learning
  2. HTML
  3. CSS
  4. BootStrap
  5. Neural Network
  6. JavaScript
  7. NodeJS
  8. Text Analytics
  9. TTS (Text-to-Speech Conversion)

oneAPI library used-

Intel® oneAPI Deep Neural Network Library (oneDNN) is an open-source performance library for deep learning applications. The library includes basic building blocks for neural networks optimized for Intel Architecture Processors and Intel Processor Graphics. It helps in increasing the application performance on Intel Architecture processors. It is used for voice training.

Documents and Presentations

Repository

https://github.com/Prajwal2001532/Smart-EduInnovators

Collaborators

3 Results

3 Results

Comments (0)