LSTMetallica: Generation drum tracks by learning the drum tracks of 60 Metallica songs by Keunwoo Choi.
From the post:
Word-RNN (LSTM) on Keras with wordified text representations of Metallica’s drumming midi files, which came from midiatabase.com.
- Midi files of Metallica track comes from midiatabase.com.
- LSTM model comes from Keras.
- Read Midi files with python-midi.
- Convert them to a text file (corpus) by my rules, which are
- (Temporal) Quantisation
- Simplification/Omitting some notes
- ‘Word’ with binary numbers
- Learn an LSTM model with the corpus and generate by prediction of words.
- Words in a text file → midi according to the rules I used above.
- Listen!
I mention this in part to inject some variety into the machine learning resources I have mentioned.
The failures of machine learning for recommendations can be amusing. For the most part when it works, they are rather dull.
Learning from drum tracks has the potential to combine drum tracks from different groups, resulting is something new.
May be fit for listening, maybe not. You won’t know without trying it.
Enjoy!