/lyrics_prediciton

Comparing Markov Chains with RNNs on text prediction

Primary LanguageJupyter Notebook

Lyrics prediction

Comparing Recurrent Neural Networks with Markov Chains in order to predict music lyrics

Currently done:

Pulled lyrics from Genius API and guided myself with Spotify Web API. Run two different kind of models and compared the modelling, the development and the outcome.

To check the dataset building, go here.

To check the markov chain analysis, go here.

To check the rnn model, go here.

To see the Medium article published in Ciencia y Datos about L. A. Spinetta, go here. (Spanish)

To see the Medium article published in Towards Data Science about Bob Dylan, go here.

This whole project has been created using Python 3 and Jupyter Notebook

I created the databases with pandas

Sources:

Predicting Logic’s Lyrics With Machine Learning by Hans Kamin

A Beginner's Guide to LSTMs and Recurrent Neural Networks

Text Predictor - Generating Rap Lyrics 📄 by Greg Surma