The following instructions are for installing and executing the roma-audio-classification case study on a linux/ubuntu machine with a GPU properly configured.
This tutorial assumes you have the CUDA driver 11.7 installed.
- Start by making sure that you have the following packages installed:
sudo apt-get install python3.9-dev
sudo apt-get install libasound2-dev
- Create a virtual environment using conda:
conda create -n roma-audio python=3.10
- Activate the environment:
conda activate roma-audio
- Install the requirements file
pip install -r requirements.txt
python main.py
Good Luck!