Exploring different strategies for the automatic generation of song lyrics with Tra-la-Lyrics



Using as starting point the architecture of a system that generates lyrics for given melodies, this paper explores possible different strategies for their automatic generation. Some important features concerning lyrics generation and some problems about different strategies are introduced. Examples of generated lyrics are shown and discussed.
The results were validated and evaluated. People were invited to answer evaluation inquiries where some lyrics could be found. The evaluation results are also a topic for discussion.


Creative Systems


13th Portuguese Conference on Artificial Intelligence (EPIA 2007), Guimarães, Portugal, December 2007

PDF File

Cited by

Year 2017 : 1 citations

 Wu, X., Du, Z., Zhong, M., Dai, S., and Liu, Y. (2017). Chinese lyrics generation using Long Short-Term Memory Neural Network. In Advances in Artificial Intelligence: From Theory to Practice, IEA/AIE 2017, volume 10351 o

Year 2014 : 2 citations

 Alfenas, D. A., Shibata, D. P., Neto, J. J., and Pereira-Barretto, M. R. (2014). Adaptive markov systems: Formulation and framework. Latin America Transactions IEEE, 12(7):1271–1277.

 Bolock, A. E. (2014). Automatic poetry generation using chr. Master’s thesis, German University in Cairo.

Year 2012 : 1 citations

 Bibi Feenaz Bhaukaurally, Mohammad Haydar Ally Didorally, Sameerchand Pudaruth. A Semi-Automated Lyrics Generation Tool for Mauritian Sega. IAES International Journal of Artificial Intelligence (IJ-AI), Vol. 1, No. 4, Dec 2012, pp. 201~213 ISSN: 2252-8938. IAES 2012.

Year 2009 : 1 citations

 Ananth Ramakrishnan A., Sankar Kuppan and Sobha Lalitha Devi. Automatic Generation of Tamil Lyrics for Melodies. Computational Approaches to Linguistic Creativity. NAACL HLT 2009