Hello world [AUDFORMAT]

Here is a hello world of using Nkululeko using dataset in audformat. This hello world is also available in Google Colab and Kaggle.

In this setup, we will use the Berlin Emodb dataset. Check the data/emodb directory for the dataset and follow the instructions below.

Change the directory to the root of the project.

# Download using wget
$ wget https://zenodo.org/record/7447302/files/emodb.zip
# Unzip
$ unzip emodb.zip
# change to Nkululeko parent directory
$ cd ..
# run the nkululeko experiment
$ python -m nkululeko.nkululeko --config tests/exp_emodb_os_xgb.ini

Then, check the results in the results directory.

You can experiment with changing some paramaters in INI file. For instance, change the type of the model to xgb or svm and see how the results change.

[EXP]
root = ./results
name = exp_emodb
[DATA]
databases = ['emodb']
emodb = ./emodb/
emodb.split_strategy = specified
emodb.train_tables = ['emotion.categories.train.gold_standard']
emodb.test_tables = ['emotion.categories.test.gold_standard']
target = emotion
labels = ['anger', 'boredom', 'disgust', 'fear', 'happiness', 'neutral', 'sadness']
[FEATS]
type = ['os']
[MODEL]
type = xgb
[PLOT]