Update README.md

This commit is contained in:
Harry Stuart 2020-01-01 15:17:14 +11:00 committed by GitHub
parent 65d92d3719
commit 726b62f304
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23

View File

@ -6,8 +6,9 @@ Since the inception of generative adversarial networks, I have been fascinated b
Music generation has many different and exciting potential applications such as: Music generation has many different and exciting potential applications such as:
Providing melody inspiration to artists - Providing melody inspiration to artists
Creating infinite, unique and free music without the need for audio file storage (for retail shops, restaurants, cafes, video games, radio stations etc.) - Creating infinite, unique and free music without the need for audio file storage (for retail shops, restaurants, cafes, video games, radio stations etc.)
GANs are already well-established in the image-processing domain, but not so much in NLP or audio-processing due to their sequential structure. After some investigaton, I learned about WaveGAN. So, I set out to adapt WaveGAN for piano in Tensorflow 2.0 using WGAN-GP as my training mechanism (as recommended by the paper). GANs are already well-established in the image-processing domain, but not so much in NLP or audio-processing due to their sequential structure. After some investigaton, I learned about WaveGAN. So, I set out to adapt WaveGAN for piano in Tensorflow 2.0 using WGAN-GP as my training mechanism (as recommended by the paper).
## What it does ## What it does