A Novel Audio Representation for Music Genre Identification in MIR
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910393789579264 |
|---|---|
| author | Kamuni, Navin Jindal, Mayank Soni, Arpita Mallreddy, Sukender Reddy Macha, Sharath Chandra |
| author_facet | Kamuni, Navin Jindal, Mayank Soni, Arpita Mallreddy, Sukender Reddy Macha, Sharath Chandra |
| contents | For Music Information Retrieval downstream tasks, the most common audio representation is time-frequency-based, such as Mel spectrograms. In order to identify musical genres, this study explores the possibilities of a new form of audio representation one of the most usual MIR downstream tasks. Therefore, to discretely encoding music using deep vector quantization; a novel audio representation was created for the innovative generative music model i.e. Jukebox. The effectiveness of Jukebox's audio representation is compared to Mel spectrograms using a dataset that is almost equivalent to State-of-the-Art (SOTA) and an almost same transformer design. The results of this study imply that, at least when the transformers are pretrained using a very modest dataset of 20k tracks, Jukebox's audio representation is not superior to Mel spectrograms. This could be explained by the fact that Jukebox's audio representation does not sufficiently take into account the peculiarities of human hearing perception. On the other hand, Mel spectrograms are specifically created with the human auditory sense in mind. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2404_01058 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | A Novel Audio Representation for Music Genre Identification in MIR Kamuni, Navin Jindal, Mayank Soni, Arpita Mallreddy, Sukender Reddy Macha, Sharath Chandra Sound Information Retrieval Machine Learning Audio and Speech Processing For Music Information Retrieval downstream tasks, the most common audio representation is time-frequency-based, such as Mel spectrograms. In order to identify musical genres, this study explores the possibilities of a new form of audio representation one of the most usual MIR downstream tasks. Therefore, to discretely encoding music using deep vector quantization; a novel audio representation was created for the innovative generative music model i.e. Jukebox. The effectiveness of Jukebox's audio representation is compared to Mel spectrograms using a dataset that is almost equivalent to State-of-the-Art (SOTA) and an almost same transformer design. The results of this study imply that, at least when the transformers are pretrained using a very modest dataset of 20k tracks, Jukebox's audio representation is not superior to Mel spectrograms. This could be explained by the fact that Jukebox's audio representation does not sufficiently take into account the peculiarities of human hearing perception. On the other hand, Mel spectrograms are specifically created with the human auditory sense in mind. |
| title | A Novel Audio Representation for Music Genre Identification in MIR |
| topic | Sound Information Retrieval Machine Learning Audio and Speech Processing |
| url | https://arxiv.org/abs/2404.01058 |