The idea behind this python machine learning project is to develop a machine learning project and automatically classify different musical genres from audio. We need to classify these audio files using their low-level features of frequency and time domain.
Music classification is a music information retrieval (MIR) task whose objective is the computational understanding of music semantics. For a given song, the classifier predicts relevant musical attributes. The retrieved information can be further utilized in many applications including music recommendation, curation, playlist generation, and semantic search.
The model trained will be used as a Single-label classification, meaning it will only label music as belonging to one of the four well known Ethiopian music genres (ቅኝቶች) namely ambasel (አምባሰል), bati (ባቲ), anchi hoye (አንቺ ሆዬ) or tizita (ትዝታ).
We have used the following algorithms to train the model in order to classify our music genres.
- KNN(K nearest neighbours)
- Random Forest
- xGBoost
- Decision Trees
- CNN (found more accurate and used in the app)
- Django for backend
- Flutter for frontend
GNU General Public License v3.0
- Tigist Wondimneh - UGR/2538/12
- Abdulkarim Getachew - UGR/7992/12
- Rediet Ferew - UGR/1415/12