An Urban sound classifier built and trained using Tensorflow 2.
The project has been organized as follows:
model/: the folder containing the trained model.requirements.txt: a text file containing the needed packages to run the repo.main.py: the script used to launch the training or testing.prepocessing.py: the code used to extract the spectrogram from the audio files and save them to npy file.test_model.py: the code used to load the saved model and classify the audio files.train_model.py: the code used to build the model and train it with the training data.
N.B: use Python 3.8
1. Clone the repo:
on your terminal, run git clone https://github.com/maky-hnou/Urban_Sounds_Classification.git
Then get into the project folder: cd Urban_Sounds_Classification/
We need to install some dependencies:
sudo apt install python3-pip libpq-dev python3-dev
2. Install requirements:
Before running the app, we need to install some packages.
- Optional Create a virtual environment: To do things in a clean way, let's create a virtual environment to keep things isolated.
Install the virtual environment wrapper:pip3 install virtualenvwrapper
Add the following lines to~/.bashrc:
export WORKON_HOME=$HOME/.virtualenvs
export PROJECT_HOME=$HOME
export VIRTUALENVWRAPPER_PYTHON=/usr/bin/python3
export VIRTUALENVWRAPPER_VIRTUALENV=~/.local/bin/virtualenv
source ~/.local/bin/virtualenvwrapper.sh
Run source ~/.bashrc
Run mkvirtualenv sound_classfier
Activate the virtual environment: workon sound_classfier (To deactivate the virtual environment, run deactivate)
- Install requirements: To install the packages needed to run the application, run
pip3 install -r requirements.txt
N.B: If you don't have GPU, or don't have Cuda and Cudnn installed, replace tensorflow-gpu by tensorflow in requirements.txt.
3- Download the dataset:
The dataset is available on Google Drive.
Download the train and test files then extract them in the repository.
4- Run the training or testing:
To run the training or the test the pre-trained model, run:
python3 main.py --mode <train/test>
Then follow the steps; you'll be asked to give the path to files and folders based on the chosen mode.