Music generation using GANs
In the previous section, we tried our hand at music generation using a very simple LSTM-based model. Now, let's raise the bar a bit and try to see how we can generate music using a GAN. In this section, we will leverage the concepts related to GANs that we have learned in the previous chapters and apply them to generating music.
We've already seen that music is continuous and sequential in nature. LSTMs or RNNs in general are quite adept at handling such datasets. We have also seen that, over the years, various types of GANs have been proposed to train deep generative networks efficiently.
Combining the power of LSTMs and GAN-based generative networks, Mogren et al. presented Continuous Recurrent Neural Networks with Adversarial Training: C-RNN-GAN4 in 2016 as a method for music generation. This is a straightforward yet effective implementation for music generation. As in the previous section, we will keep things simple and focus only...