Skip to content

kkothari2001/twAIn

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

93 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

twAIn

An AI that generates amazing stories!

Team Name : GoogleKarnaHamseSikho

Screenshot alt text

👩‍💻 Team members

  • Ravi Maurya 🦉
  • Kush Kothari 😎
  • Shreyas Penkar 🦊
  • Nikheel Indanoor 🦅

👨‍🏫 Mentors

  • Mehdi Patel
  • Anuj Raghani
  • Owais Hetavkar

📝 Description

twAIn is an Artificial Intelligence that can write amazing stories. twAIn is powered by LSTM and GPT-2 Models. It is trained on 20+ famous and compelling novels of various Genres. twAIn can be used as a Web App and a Mobile App.
Give some starting words and twAIn will write an amazing story for you. Currently the genres supported are Adventure, Horror and Mystery. The user has the option to listen, as twAIn reads the story aloud for you. Also there are 45 sample stories preloaded so that you can see the performance of that model.

🔗 Links

🤖 Technology Stack

  • Front-end - HTML, CSS, Javascript
  • Back-end - Python, Flask (Flask is a lightweight WSGI web application framework.)
  • Models - LSTM, Open AI's GPT-2
  • Mobile App - Flutter

🛠️ Project Setup

For Web

  1. Clone the repository
git clone https://github.com/kkothari2001/twAIn
  1. Download the models from here (You can download single genre or all genres)
  2. After downloading the models, add the models in this path
twAIn/WebApp/checkpoint/<genreNameFolders>

Eg. twAIn/WebApp/checkpoint/adventure
                           /horror
                           /mystery
  1. Now, Install all the python packages from WebApp/requirements.txt in a Virtual Environment.

For Mobile

  1. Follow All the Instructions of Web First.
  2. Make sure you have Flutter installed on your PC.
  3. Go to MobileApp/twain_app and run flutter pub get to install all dependencies.
flutter pub get
  1. Open file MobileApp/twainapp/lib/utils/settings.dart. In this file set MODE to **_production**, and set IP_ADD to your local IP Address. (So that you can test it locally).
  2. Install the app on your physical Mobile Device or Emulator.

🏃‍♀️ Usage

After setting up for Web and Mobile follow the below instructions to get it running.

For Web

  1. Goto twAIn/WebApp & Execute this to run the Web App
python app.py
  1. Open localhost:8000/ in your browser to use twAIn.

For Mobile

  1. Goto twAIn/WebApp & Execute this and keep it running in the background.
python app.py
  1. Make sure your PC and Mobile is connected on same Wifi Network. (Or your PC is connected to your Mobile's Hotstop) & You can followed the Project Setup for Mobile App above.
  2. Open the app in your Mobile Phone to use twAIn.

⚡ Applications

This project generates stories based on the input provided by the user. There are various genres for the user to select which includes Adventure ,Horror,Mystery. It can also convert the generated story into audio.

For Web

  • Type "help" to get a help-page on your screen.
  • Type "sample" to generate a sample story.
  • Type "download" to download the generated story as a text file.

The stories to be generated take some time:-

  • short story (250 chars) = max 2 mins
  • medium story (550 chars) = max 4 mins
  • long story (900 chars) = max 6 mins

If you dont wish to wait so much time you can type "sample" and press Enter or click Submit (Web) to get a sample story of that genre and of a specific story length. There are 45 sample stories preloaded and take about 3 seconds to load.

🔮 Future Scope

  • We could add more genres
  • Try to improve our LSTM model to maximize the results for story generation.
  • We could also do a model for generating poems and other creative writing templates.
  • We could also include a sentimental analysis model which will tell how good a story is (by a small survey we can know the sentiments of a user towards a particular story).

🖼 Screenshots

Web App Mobile App

About

An AI that creates incredible stories!

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 4

  •  
  •  
  •  
  •