Enhanced capsule neural network with advanced triangulation topology aggregation optimizer for music genre classification
Abstract Music genres classification poses a formidable challenge as it necessitates capturing the intricate and varied characteristics of musical signals. In this study, an innovative approach is presented to classify the music genres using the Capsule Neural Network (CapsNet). The CapsNet model op...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-01-01
|
| Series: | Scientific Reports |
| Subjects: | |
| Online Access: | https://doi.org/10.1038/s41598-024-83577-z |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Abstract Music genres classification poses a formidable challenge as it necessitates capturing the intricate and varied characteristics of musical signals. In this study, an innovative approach is presented to classify the music genres using the Capsule Neural Network (CapsNet). The CapsNet model optimized by an advanced version of Triangulation Topology Aggregation Optimizer (ATTAO). CapsNet effectively preserves the spatial and hierarchical information of the input data, while ATTAO efficiently optimizes the parameters of CapsNet. The proposed method applied to two extensively utilized datasets, namely GTZAN and Ballroom, and compare its performance against several cutting-edge techniques. Here we show that based on the experimental findings, unequivocally demonstrate that our method outperforms others in different terms, thereby showing its efficacy and resilience in music genre recognition. |
|---|---|
| ISSN: | 2045-2322 |