training_setup.txt

-To get metrolyrics data:
First download the csv from here: https://www.kaggle.com/gyani95/380000-lyrics-from-metrolyrics
Unzip and save the csv file at ./data/metrolyrics/lyrics.csv

-If you want to use your own lyrics dataset, you can download the profanity filtering data here:
To get the list of profane words to filter:
`wget https://github.com/zacanger/profane-words/archive/master.zip`
`unzip master.zip`
`rm master.zip`
`mv profane-words-master data/profane-words`

-To download librispeech data:
`wget http://www.openslr.org/resources/12/train-clean-100.tar.gz`
Unzip and move to data/librispeech/LibriSpeech/train-clean-100

`wget http://www.openslr.org/resources/12/dev-clean.tar.gz`
Unzip and move to data/librispeech/LibriSpeech/dev-clean

-Download librispeech alignments from https://github.com/CorentinJ/librispeech-alignments

-Download 100d GloVe vectors from: http://nlp.stanford.edu/data/glove.6B.zip