Towards the Generation of Musical Explanations with GPT-3

Open AI’s language model, GPT-3, has shown great potential for many NLP tasks, with applications in different domains. In this work we carry out a first study on GPT-3’s capability to communicate musical decisions through textual explanations when prompted representation of piece music. Enabling dialogue human-AI music partnerships is an important step towards more engaging and creative interactions. Our results show that GPT-3 lacks the necessary intelligence really understand decisions. A major barrier reach better performance lack data includes process carried by artists pieces. We believe such resource would aid understanding collaboration AI systems.