2022 · 12 citations · 6 references
ComputingComputer ScienceMachine LearningMusic AnalysisNatural Language ProcessingClassificationDeep LearningGenre ClassificationLyrics AnalysisMusical GenresPattern RecognitionMachine Learning For MusicMusic ClassificationMusic Genre Classification SystemsSong Classification
Music is one of the most important factors of human lives. People express their inner thoughts, emotions and feelings with the combination of both lyrics and musical instruments. With the passage of time, languages have also evolved and the structure of the languages have also changed. By listening to music, we can understand which genre the song belongs to. So, we have come up with an idea of classifying genre based on the music lyrics. There are few works where the researchers worked on genre based lyrics classification. However, those works were based on different languages such as English, Hindi and Spanish. As a result, we have got an opportunity to work on Bangla language. In this paper, we have used different machine learning (ML) models and a custom CNN model to classify Bangla Lyrics on “Bangla Song Lyrics” dataset which is available on Kaggle. Our proposed CNN model achieved highest accuracy of 69.36% with an F1-score of 69.17% among all models.
6
Scikit-learn: Machine Learning in Python
PedregosaFabian, VaroquauxGaël, GramfortAlexandre et al. · Journal of Machine Learning Research · 2011 · 8.2K citations
AdaBoost typical Algorithm and its application research
TU Cheng-sheng, Liu Huacheng, Bing Xu · MATEC Web of Conferences · 2017 · 161 citations · Full text