EPISODE · Dec 31, 2020 · 0 MIN
AI Bach GANS
from OPEN MUSIC PERFORMANCE · host Start by Keizo Akutagawa
This music was composed by AI with deep learning algorithm which used standalone Keras with a Tensorflow 1.14 backend. It was trained by Bach's four-tone chorus dataset of 229 songs and played on LogicPro X. Using the method of extracting the characteristics of the composer by calculating the probability distribution of the pitch and length of the note that comes after each note. The official code repository for examples in the O'Reilly book 'Generative Deep Learning' https://learning.oreilly.com/library/view/generative-deep-learning/9781492041931/ https://www.amazon.com/Generative-Deep-Learning-Teaching-Machines/dp/1492041947/ref=sr_1_1
What this episode covers
This music was composed by AI with deep learning algorithm which used standalone Keras with a Tensorflow 1.14 backend. It was trained by Bach's four-tone chorus dataset of 229 songs and played on LogicPro X. Using the method of extracting the characteristics of the composer by calculating the probability distribution of the pitch and length of the note that comes after each note. The official code repository for examples in the O'Reilly book 'Generative Deep Learning' https://learning.oreilly.com/library/view/generative-deep-learning/9781492041931/ https://www.amazon.com/Generative-Deep-Learning-Teaching-Machines/dp/1492041947/ref=sr_1_1
NOW PLAYING
AI Bach GANS
No transcript for this episode yet
Similar Episodes
No similar episodes found.