Skip to content

BERTunes-Classifier: Developed a music genre classification system achieving 93% accuracy using machine learning algorithms and textual analysis of song lyrics.

Notifications You must be signed in to change notification settings

tejaswidabas123/BERTunes-Classifier

Repository files navigation

BERTunes-Classifier

MUSIC GENRE CLASSIFICATION

• Developed a music genre classification system utilizing various machine learning algorithms, including logistic regression, Naïve Bayes, linear SVM, polynomial SVM, Gaussian SVM, KNN, and CNN.

• Successfully classified 10 different music genres based on audio features, including rhythm, tempo, and pitch with an accuracy of 93%.

• Utilized Speech Recognition to extract text from songs and Employed transformer-based models, including BERT, Glove & Word2vec to analyze and classify song lyrics based on themes, emotions, and other textual characteristics.

• Conducted extensive exploratory data analysis (EDA) and feature engineering to identify the most important audio and textual features for classification and developed comprehensive data pre-processing pipelines to prepare audio and text data.

Dashboard Image

About

BERTunes-Classifier: Developed a music genre classification system achieving 93% accuracy using machine learning algorithms and textual analysis of song lyrics.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published