Tamil Filmy Music Genre Classifier using Deep Learning Algorithms Christina Gnanamani 1 and Shanthini Pandiaraj 2 1 Department of ECE,Coimbatore Institute of Technology, TN, India https://orcid.org/0000-0003-3797-6165 2 Department of ECE, Karunya Institute of Technolog and Sciences, TN, India https://orcid.org/0000-0003-1743-9661 Abstract—The objective of this work is to classify the genre of the given Tamil music using CNN algorithm. Music plays an important role in human’s life. Due to human evolution, the way of listening music and music catalogues also get advanced. So, this classifier technique helps the user to find the frequently used genre of that particular Tamil music. There are billions of songs in this world, so it is very difficult to classify the Tamil music genre of the songs by manual process for music streaming companies. Thus an accurate and predefined classifier is required, which can be succeed with deep learning using convolution neural network (CNN) algorithms. In this project, Jupyter notebook is used to compile the python code. The Machine is trained using GTZAN dataset which consist of 1000 audio tracks with 10 different types of Tamil music genres. To predict the specific Tamil music genre of the given audio file the feature of both the train and test data will be extracted and classification will be done with Convolution neural network algorithm and the output will be displayed. Index Terms— CNN algorithms, Jupyter notebook, GTZAN datasets. I. INTRODUCTION Music plays an important role in every human’s life. Earlier, people used to listen music only for their pleasure and to express their feelings, but nowadays people listen to songs for pleasure and they also analyze the technical terms behind that song, including the instruments used and the singer. Music is universal and composers’ worldwide plagiarize or use foreign instruments to enrich their music and provide variety for listeners. We all are familiar with the elements of the music such as Melody, Folk, and Western. But Tamil music genre has songs from genres such as Jazz, Blues, Hip hop, Metal and so on. People have loved and listened to these types of songs without knowing their music genre. There is also several music streaming companies which act as a best entertaining platform for music lovers. As a listener listens to a particular song on repeat mode, the app identifies the genre and suggests songs in the same genre. This can be suggested with our Tamil music genre Classifier. Tamil music genre classifier is a model which is able to classify frequently used genre of the particular song. This can be done with the help of Convolution Neural network algorithm (CNN) which is a deep learning approach. Deep learning is a subset of Machine learning which deals with large amount of data set to train the machine. The 2 GTZAN data set consist of train and test data. There are 1000 audio files with 10 different types of genre each genre consist of 100 audio files. Totally 1000 audio files are there, each audio is 30 seconds in duration and they are in ‘.WAV’ format. The features of the training data set are extracted and they are trained Grenze ID: 01.GIJET.8.1.110 © Grenze Scientific Society, 2022 Grenze International Journal of Engineering and Technology, Jan Issue