Persian MusicGen: A Large-Scale Dataset and Culturally-Aware Generative Model for Persian Music

TL;DR

This paper introduces a large-scale Persian music dataset and adapts a state-of-the-art generative model to produce culturally authentic Persian music, addressing the gap in non-Western music generation.

Contribution

It provides the first extensive Persian music dataset and demonstrates how to fine-tune a generative model for culturally-specific music synthesis.

Findings

Fine-tuned model produces music aligning with Persian stylistic conventions.

Dataset captures diverse Persian musical styles and cultural nuances.

Model evaluation shows improved semantic alignment with style tags.

Abstract

Persian music, with its unique tonalities, modal systems (Dastgah), and rhythmic structures, presents significant challenges for music generation models trained primarily on Western music. We address this gap by curating the first large-scale dataset of Persian songs, comprising over 900 hours high-quality audio samples across diverse sub-genres, including pop, traditional, and contemporary styles. This dataset captures the rich melodic and cultural diversity of Persian music and serves as the foundation for fine-tuning MusicGen, a state-of-the-art generative music model. We adapt MusicGen to this domain and evaluate its performance by utilizing subjective and objective metrics. To assess the semantic alignment between generated music and intended style tags, we report the proportion of relevant tags accurately reflected in the generated outputs. Our results demonstrate that the fine-tuned model produces compositions that more align with Persian stylistic conventions. This work introduces a new resource for generative music research and illustrates the adaptability of music generation models to underrepresented cultural and linguistic contexts.