Paper Link: Synthetic Speech Classification: IEEE Signal Processing Cup 2022 challenge
Download the zip file of the data set from here.
- It is recommended to use Anaconda to create a virtual environment before proceeding to run code.
- It is recommended to stick to Python 3.7.11
- The following are the main requirements:
torch
torchvision
torchaudio
pytorch-lightning
scikit-learn
numpy
pandas
scikitplot
matplotlib
librosa
soundfile
wandb (for logging)
- The deep learning parts of the codebase was tested on CUDA 10.1 and cuDNN 7.6.5.
- Mahieyin Rahmun - [email protected]
- Zulker Nayeen Nahiyan - [email protected]
- Sadia Khan - [email protected]
- Rafat Hasan Khan - [email protected]
- Rakibul Hasan Rajib - [email protected]
- Mir Sayad Bin Almas - [email protected]
- Dr. Sakira Hassan - [email protected]
- Tanjim Taharat Aurpa - [email protected]