Preprocessing Song Dataset for Melody Generation - Pt. 2
Valerio Velardo - The Sound of AI via YouTube
Overview
Save Big on Coursera Plus. 7,000+ courses at $160 off. Limited Time Only!
Dive into the second part of a tutorial on preprocessing song datasets for melody generation using LSTM networks. Explore techniques for encoding songs as time series, a crucial step in preparing data for neural network training. Learn to convert musical notes into a suitable representation, choose appropriate symbols, and translate the data into strings. Discover how to update preprocessing functions and save the processed songs. Test the implemented code to ensure proper functionality. This video is part of a comprehensive series on generating melodies with LSTM networks, providing practical insights for building effective artificial neural networks for music creation.
Syllabus
Introduction
encodesong function
convert notes to time series
choose symbol
review
translating to strings
updating preprocessing function
saving the song
testing the code
Taught by
Valerio Velardo - The Sound of AI