MSc_LangTech_NKUA_DSP

Code for MSc_LangTech_NKUA_DSP courses.

0. If using windows, install git for windows. https://git-scm.com/download/win In Mac, you might need to install git, there are several workarounds: https://git-scm.com/book/en/v2/Getting-Started-Installing-Git In Linux, it's already there...

1. Create a folder for all your repositories - make sure you know how to navigate there. E.g. create the directory "repos" in "Documents" - full path would be: Documents/repos (assuming you're in the home folder of the current user).

2. Only for the first time you need to get the files in a particular computer, use clone. -- 2.1 If in windows, open the "Git Bash", else open a terminal and navigate to the repos folder (e.g. type "cd Documents/repos"). -- 2.2 Clone the repository as "git clone https://github.com/maximoskp/MSc_LangTech_NKUA_DSP.git".

3. Any time you want to refresh the content of the repository, navigate to its folder, created in "repos", and type "git pull".

• Basic UNIX commands - to be used in the Git console or in Terminal (if on Mac or Linux): http://www.econ.ucla.edu/TApages/wan/basic_commands.html

- Install Anaconda

Download and install Anaconda in your machine (Windows / macOS / Linux):

https://www.anaconda.com/products/individual

Open Anaconda navigator and create a new environment using the latest version of python. Choose a name for your environment (e.g. name it "audio_dsp").

In the "created environment", install "Jupyter Notebook".

Open Jupyter Notebook - preferably from terminal.

auto open in chromium / or any other browser:

jupyter notebook --generate-config

code jupyter_notebook_config.py # or nano, or whatever browser

change the line

# c.NotebookApp.browser = ''

to

c.NotebookApp.browser = 'C:/path/to/your/chrome.exe %s' - Caution: backslashes, regardless of windows Caution: %s at the end

Change color theme:

conda install -c conda-forge jupyterthemes

or

pip install jupyterthemes

jt -t chesterish

- Create environments

More information here: https://conda.io/projects/conda/en/latest/user-guide/tasks/manage-environments.html

Initially, it is ok to have one environment for all courses or applications, e.g., you might name it "general".

- Install Spyder

The best way to do it, if you don't have Mac M1 computer, is through the Anaconda navigator or the command "pip install spyder" in the command line. If you have a Mac M1, your best bet is to use the standalone app.

- Install libraries

Install interactive extension pip install ipywidgets pip install jupyterlab jupyter labextension install @jupyter-widgets/jupyterlab-manager

You might also need to conda install matplotlib

• Useful libraries

to install with pip install numpy matplotlib librosa