Composing Bach Chorales Using Deep Learning • Feynman Liang • GOTO 2019

This presentation was recorded at GOTO Copenhagen 2019. #GOTOcon #GOTOcph

Feynman Liang - Creator of BachBot

ABSTRACT
Can musical creativity, something believed to be deeply human, be codified into an algorithm?
While most music theorists are hesitant to claim a "correct" algorithm for composing music like Bach, recent advances in machine learning and computational musicology may help us reach an answer.
In this talk, we describe BachBot: an artificial intelligence which uses deep learning and long short term memory (LSTM) to compose music in the style of Bach. We train BachBot on all known Bach chorale harmonisations and carry out the largest musical Turing test to date. Our results show that the average listener can distinguish BachBot from real Bach only 5% better than random guessing, suggesting that algorithmic composition of Bach chorales is more closed (as a result of BachBot) [...]

Download slides and read the full abstract here:

#BachBot #BachChorals #AI #ML #DataScience #DeepLearning

Looking for a unique learning experience?

SUBSCRIBE TO OUR CHANNEL - new videos posted almost daily.

[pauljarski]

It occurs to me that they're approaching this problem backwards. Why train a computer to guess what Bach would have written, when in fact Bach was simply applying an algorithm, which he never wrote down but which nevertheless exists and is taught, somewhat obliquely, to every serious classical musician? If the algorithm exists, you can simply code it. You don't need this roundabout way.

I would like to see some demonstration that musical informatics is not a waste of time. Can anyone help me?

[Mike-cptj]

Fixed your title: *Composing Garbage that Sounds Nothing Like Bach Using Computer*

[playbach]

Excuse me but this is so far from sounding anything like Bach.
The weak harmony and horrible voice leading makes a really bad realisation..
But keep trying.