Iranian Modal Music (Dastgah) detection using deep neural networks

TL;DR

This study applies deep neural networks to recognize Iranian modal music (Dastgah) with high accuracy, highlighting the effectiveness of MFCC features and autoencoder-inspired architectures in distinguishing highly correlated musical categories.

Contribution

Introduces a deep learning approach, BiLGNet, for Iranian modal music classification, demonstrating superior accuracy and robustness over other models in non-western music genre detection.

Findings

BiLGNet achieved 92% accuracy in Dastgah detection.

MFCC features outperform other sound representations.

Bidirectional recurrent networks are most effective for this task.

Abstract

Music classification and genre detection are topics in music information retrieval (MIR) that many articles have been published regarding their utilities in the modern world. However, this contribution is insufficient in non-western music, such as Iranian modal music. In this work, we have implemented several deep neural networks to recognize Iranian modal music in seven highly correlated categories. The best model, BiLGNet, which achieved 92 percent overall accuracy, uses an architecture inspired by autoencoders, including bidirectional LSTM and GRU layers. We trained the models using the Nava dataset, which includes 1786 records and up to 55 hours of music played solo by Kamanche, Tar, Setar, Reed, and Santoor (Dulcimer). We considered Multiple features such as MFCC, Chroma CENS, and Mel spectrogram as input. The results indicate that MFCC carries more valuable information for detecting Iranian modal music (Dastgah) than other sound representations. Moreover, the architecture inspired by autoencoders is robust in distinguishing highly correlated data like Dastgahs. It also shows that because of the precise order in Iranian Dastgah Music, Bidirectional Recurrent networks are more efficient than any other networks that have been implemented in this study.