A research of using multi-modal for Vietnamese music authors classification task.

Also this is our graduation thesis at HCMUS. Graduation year: 2023.

To receive full dataset, notebooks please contact: [email protected]

1. Dataset
   • Data collecting system
   • Dataset description
2. Models' Description
3. Research Result
4. Training Notebooks

Our full data construction phase:

Framework: Scrapy (version 2.8)
Language: Python (version 3.9)
Source code: CSN Crawler

Collecting process:

Our dataset includes 15 Vietnamese authors and 50 songs each, except the last author. The train, validation and test set account for 60%, 20% and 20% of the full dataset respectively.

CSV columns definition

Contact for full access to our dataset.

Proposed approaches:

Self-build from Cross-Modal Attention MultiModal of Krishna D N.

Our multimodal approach

Metrics score:

Confusion matrices:

Our implementing and training phase mostly on Google Colaboratory mechanism. Therefore, there was no specific version of local training setups for these models.

Contact for detailed notebooks.

In this repository, we provide 4 notebooks with 4 main phases:

• Load & visual datasets.
• Initalize Dataset class and Model class.
• Configure hyper-parameters, specific settings & Train model.
• Visualize the loss curves and metrics.

MIT License

Copyright (c) 2023 PhmNm

Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.