Training a Long-Short Term Memory Network for Melody Generation
Valerio Velardo - The Sound of AI via YouTube
Overview
Build and train an LSTM network capable of generating folk songs using Tensorflow and Keras in this 31-minute tutorial video. Explore the Keras functional API while learning preprocessing techniques, creating training sequences, and constructing a neural network model. Follow along as the instructor guides you through adding input layers, nodes, and upper layers before compiling, training, and saving the model. Conclude by testing the newly created melody generator and gain insights into the process of developing AI-powered music composition tools.
Syllabus
Introduction
Preprocessing
Training
Training Sequences
Building the Network
Building the Model
Input Layer
Adding Nodes
Building the Upper Layer
Compile the Model
Train the Model
Save the Model
Testing the Model
Conclusion
Taught by
Valerio Velardo - The Sound of AI