A pipeline to build a dataset from your own music library and use it to fill the missing genres
Read the article on Medium
Required install:
eyed3
sox --with-lame
tensorflow
tflearn
- Create folder Data/Raw/
- Place your labeled .mp3 files in Data/Raw/
To create the song slices (might be long):
python main.py slice
To train the classifier (long too):
python main.py train
To test the classifier (fast):
python main.py test
- Create folder Input/Raw/
- Place your unlabeled .mp3 files in Input/Raw/
To create song slices of song to predict:
python main.py sliceInput
To predict the classifier's output:
python main.py predict
- Most editable parameters are in the config.py file, the model can be changed in the model.py file.
- Pipeline to label new songs with the model is given below