LGSDASJun 1, 2023

Transfer Learning for Underrepresented Music Generation

arXiv:2306.00281v12 citationsh-index: 16
Originality Synthesis-oriented
AI Analysis

This addresses the challenge of generating underrepresented music genres for music AI applications, though it is incremental as it builds on existing MusicVAE methods.

The paper tackled the problem of generating underrepresented music genres by adapting MusicVAE to Iranian folk music using a combinational creativity transfer learning approach, resulting in efficient adaptation to this out-of-distribution genre.

This paper investigates a combinational creativity approach to transfer learning to improve the performance of deep neural network-based models for music generation on out-of-distribution (OOD) genres. We identify Iranian folk music as an example of such an OOD genre for MusicVAE, a large generative music model. We find that a combinational creativity transfer learning approach can efficiently adapt MusicVAE to an Iranian folk music dataset, indicating potential for generating underrepresented music genres in the future.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes