Skip to content

Machine Learning generative text system with Twitter bot functionality

Notifications You must be signed in to change notification settings

RyanJennings1/CC-Song-Generator

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

31 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Introduction

This is the repo for Auto Song Generation of Irish Folk songs.
This is part of my COMP47410 Computational Creativity Assignment

📒 Index

🔰 About

This project is part of the final assignment for COMP47410 Computational Creativity. The repo contains a machine learning model that takes a folder of lyrics and outputs generated text before posting it on twitter. The lyrics directory can be swapped with any text type data to train then generated.

🔌 Installation

$ git clone https://github.com/RyanJennings1/CC-Song-Generator.git

Install the project before use to include all the dependencies:

  • tensorflow==1.13.1
  • tweepy
  • pyenchant
  • nltk
$ python3 setup.py install

Also Required A Twitter developer account is needed to get Twitter API secrets which can be used to make the bot post to twitter.

Logins for making a connection to Twitter are brought in from user's shell environment values twitter_api_key, twitter_api_key_secret, twitter_access_token and twitter_access_token_secret

☀️ Usage

After installation and being built just run the binary

$ ./bin/ccsonggenerator --help
usage: ccsonggenerator [-h] [--train] [--run] [--analyse] [--version]

optional arguments:
  -h, --help     show this help message and exit
  --train, -t    Train the model on lyrics data
  --run, -r      Run the model to generate an output
  --analyse, -a  Analyse existing tweets published
  --version, -v  Print version

🔧 Development

If anybody wants to contribute to this project or fork it feel free.

🌟 Credit/Acknowledgment

Ryan Jennings
burliEnterprises

About

Machine Learning generative text system with Twitter bot functionality

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages