Digitalization of music is the new trend, and preferences of individuals are highly rated. Millions of songs are being streamed in the music applications. The companies providing these services need to sort and arrange a wide range of music tastes for all of its users. On top of that, fresh music from various artists in a wide spectrum of genres are popping up every day. To keep track of all this, a classification system can be handy. So, we propose an RNN based model based on Natural Language processing to classify the songs based on their lyrics into different genres [1]. Additionally, this tool can be handy to the music lovers for quickly identifying which genre a particular song belongs to. In this paper, we apply Long Short Term Memory (LSTM) model with both Universal Serial Embedder (USE) and Bert embedders. A comparative study is performed to understand which combination of models works based to classify the genres based on lyrics. From our results, on the basis of accuracy of the model, we found that USE embedder with LSTM [2] gives a slightly better performance than Bert embedder. The LSTM model with USE embedding gave the highest accuracy of 83.42% when trained over a range of five folds.
Skip Nav Destination
Article navigation
5 June 2024
INTERNATIONAL CONFERENCE ON RESEARCH IN SCIENCES, ENGINEERING, AND TECHNOLOGY
28–29 November 2022
Warangal, India
Research Article|
June 05 2024
Song lyrics genre detection using RNN
Syed Nawaz Pasha;
Syed Nawaz Pasha
a)
1
School of Computer Science &Artificial Intelligence, SR University
, Warangal, Telangana, India
a)Corresponding author: [email protected]
Search for other works by this author on:
Dadi Ramesh;
Dadi Ramesh
1
School of Computer Science &Artificial Intelligence, SR University
, Warangal, Telangana, India
Search for other works by this author on:
Sallauddin Mohmmad;
Sallauddin Mohmmad
1
School of Computer Science &Artificial Intelligence, SR University
, Warangal, Telangana, India
Search for other works by this author on:
Shabana;
Shabana
2
Sumathi Reddy Institute of Technology for Women
, Warangal, Telangana, India
Search for other works by this author on:
D. Kothandaraman;
D. Kothandaraman
1
School of Computer Science &Artificial Intelligence, SR University
, Warangal, Telangana, India
Search for other works by this author on:
T. Sravanthi
T. Sravanthi
2
Sumathi Reddy Institute of Technology for Women
, Warangal, Telangana, India
Search for other works by this author on:
a)Corresponding author: [email protected]
AIP Conf. Proc. 2971, 020055 (2024)
Citation
Syed Nawaz Pasha, Dadi Ramesh, Sallauddin Mohmmad, Shabana, D. Kothandaraman, T. Sravanthi; Song lyrics genre detection using RNN. AIP Conf. Proc. 5 June 2024; 2971 (1): 020055. https://doi.org/10.1063/5.0195902
Download citation file:
Pay-Per-View Access
$40.00
Sign In
You could not be signed in. Please check your credentials and make sure you have an active account and try again.
21
Views
Citing articles via
Inkjet- and flextrail-printing of silicon polymer-based inks for local passivating contacts
Zohreh Kiaee, Andreas Lösel, et al.
Effect of coupling agent type on the self-cleaning and anti-reflective behaviour of advance nanocoating for PV panels application
Taha Tareq Mohammed, Hadia Kadhim Judran, et al.
Design of a 100 MW solar power plant on wetland in Bangladesh
Apu Kowsar, Sumon Chandra Debnath, et al.
Related Content
Classification machine learning with chromagram and word embedding feature
AIP Conf. Proc. (January 2025)
Hit songs prediction: A review on machine learning perspective
AIP Conf. Proc. (January 2024)
Audio classification based on genre
AIP Conference Proceedings (May 2022)
Comparison of improved support vector machine and multilayer perceptron for the prediction of music genre with improved classification accuracy
AIP Conf. Proc. (November 2023)
Exploring cross-genre performance for age-based fiction classification models
AIP Conf. Proc. (February 2023)