underwater_acoustic_target_classification
Implementation of AST and MelGAN for Underwater Acoustic Target Classification
https://github.com/devichand579/underwater_acoustic_target_classification
Science Score: 44.0%
This score indicates how likely this project is to be science-related based on various indicators:
-
✓CITATION.cff file
Found CITATION.cff file -
✓codemeta.json file
Found codemeta.json file -
✓.zenodo.json file
Found .zenodo.json file -
○DOI references
-
○Academic publication links
-
○Academic email domains
-
○Institutional organization owner
-
○JOSS paper metadata
-
○Scientific vocabulary similarity
Low similarity (8.1%) to scientific vocabulary
Repository
Implementation of AST and MelGAN for Underwater Acoustic Target Classification
Basic Info
Statistics
- Stars: 1
- Watchers: 1
- Forks: 0
- Open Issues: 0
- Releases: 0
Metadata Files
README.md
A Novel Approach to Underwater Acoustic Target Classification with MelGAN and Audio Spectrogram Transformer
This work implements Audio Spectrogram Transformer on ShipsEar Database (A private underwater vessel noise database) which serves as the benchmark for various models on underwater noise classification. Training Dataset was expanded by generation of synthetic audio samples using MelGAN.
- Refer to the document for further details 📖
Data Format
MelGAN
Architecture of MelGAN , Generator on the left and Discriminator on the right.
Audio Spectrogram Transformer
Architecture of AST.
Code
```bash -data_sampler.ipynb -- This notebook contains code for converting raw audio files of the database into suitable formats and lengths for training AST
-melgan.ipynb -- This notebook contains the implementation of MelGAN for generating synthetic audio samples. This notebooks has to be run for each class of the database seperately.
-train_test.ipynb -- This notebook splits both the real and synthetic data into train, validation and test splits. we suggest to run this code for each class of the database seperately for mitigating class imbalance.
-ast.ipynb -- This notebook contains the implementation of AST. ```
Contribution and Funding
- This work is done in collaboration with Indian Institute of Technology Palakkad.
- IIT Palakkad Technology Hub has funded the project.
References
if you find this work useful, please cite this repository:
bibtex
@software{Budagam_A_Novel_Approach_2024,
author = {Budagam Devichand and Manikandan Sabarimalai},
month = may,
title = {{A Novel Approach to Underwater Acoustic Target Classification with MelGAN and Audio Spectrogram Transformer}},
url = {https://github.com/devichand579/Underwater_acoustic_target_classification},
year = {2024}
}
Owner
- Name: Devichand
- Login: devichand579
- Kind: user
- Repositories: 1
- Profile: https://github.com/devichand579
Citation (CITATION.cff)
cff-version: 1.2.0 message: "If you use this repository, please cite it as below." authors: - family-names: "Budagam" given-names: "Devichand" - family-names: "Manikandan" given-names: "Sabarimalai" title: "A Novel Approach to Underwater Acoustic Target Classification with MelGAN and Audio Spectrogram Transformer" date-released: 2024-05-16 url: "https://github.com/devichand579/ROLEBENCH"
GitHub Events
Total
- Watch event: 1
Last Year
- Watch event: 1