|Abstract||As music collections are growing up every day, it becomes necessary to keep them organized. Therefore, one critical block of an 'intelligent' system for doing that should deal with the concept of music similarity. From a computational point of view, this concept has been assessed from many angles, being one of the most important the purely audio content-based similarity systems. Traditionally, they have relied in methods that ignored the temporal evolution of the music audio characteristics. It is the goal of this thesis to provide some insights into the benefits of considering temporal sequences representing the musical content of an audio signal. We do that while focusing to an application that have been increasing his popularity along these last few years, as it provides a direct and objective way for evaluating music similarity cover song identification. A cover song (or simply cover) is a new version (performance, rendition or recording) of a previously recorded song. After making an extensive literature summary on related techniques, methods and systems for music similarity (with a special emphasis on cover detecting systems), and presenting our evaluation methodology, we perform several experiments on cover song identification based on state-of-the-art methods (cross-correlation and Dynamic Time Warping). We also study the blocks of these methods that can report more benefits to the final performance of the system and we make some interesting improvements on them (such as considering different PCP distances, beat tracking algorithms, key transposition methods), apart from assessing the intrinsic algorithms' parameters. Furthermore, we propose a new method for determining the similarity between tonal sequences and, therefore, to cover songs. This one is based on a novel HPCP similarity measure, and on a newly developed Dynamic Programming local alignment technique. Results confirm that the performance of the proposed system is significantly superior to the other implemented ones. Along the thesis we keep discussing important details found during the experiments done and future directions to take for accomplishing the chosen task.