GANs & Reels: Creating Irish Music using a Generative Adversarial Network
This work addresses music generation for Irish traditional music, but it is incremental as it adapts existing GAN architectures to a specific domain without major methodological breakthroughs.
The authors tackled algorithmic melody generation by using a generative adversarial network without recurrent components, specifically applying a DC-GAN with dilated convolutions to create Irish traditional reel melodies, achieving results that capture long-range dependencies in fixed-length forms.
In this paper we present a method for algorithmic melody generation using a generative adversarial network without recurrent components. Music generation has been successfully done using recurrent neural networks, where the model learns sequence information that can help create authentic sounding melodies. Here, we use DC-GAN architecture with dilated convolutions and towers to capture sequential information as spatial image information, and learn long-range dependencies in fixed-length melody forms such as Irish traditional reel.