In the previous chapter, we learned about generating text by going through a novel. In this section, we will learn about generating audio from a sequence of audio notes.
Generating music using deep learning
Getting ready
A MIDI file typically contains information about the notes and chords of the audio file, whereas the note object contains information about the pitch, octave, and offset of the notes. The chord object contains a set of notes that are played at the same time.
The strategy that we'll adopt to build a music generator is as follows:
- Extract the notes present in audio file
- Assign a unique ID for each note.
- Take a sequence of 100 historical notes, and the 101st note shall be the output.
- Fit an LSTM model...
