Query by music segments: an efficient approach for song retrieval

Scalable multi-feature index structure for music databases

“…Recently, with the rapid progress in digital representations of music data, the methods for efficiently managing a music database are receiving more attention. There are more and more increasingly attractive investigations for retrieving music collections such as the Query by Rhythm by Chen et al [4], Query by Music Segments by Chen et al [5], Multi-Feature Index Structures by Lee and Chen [18], Non-Trivial Repeating Pattern Discovering by Liu et al [22], Linear Time for Discovering Non-Trivial Repeating Patterns by Lo and Lee [27], Approximate String Matching Algorithm by Liu et al [20], Key Melody Extraction and N-note Indexing by Tseng [34], Melodic Matching Techniques by Uitdenbogerd and Zobel [37], Numeric Indexing for Music Data by Lo and Chen [23,24] and more in [1][2][3]12,17,21,23,26,28,31,35,36].…”

An Efficient Numeric Indexing Technique for Music Retrieval System

“…No matter how the pre-existing methods [1][2][3][4] work, they view the music object as a string, and the processes of indexing and retrieval are by means of text-related techniques.…”

Finding maximum-length repeating patterns in music databases

“…Finally, when finished with the traversal, all candidates that are linked to the initially found MLRP (i.e., those with length equal to the found maximum found length for |C|) are examined so as to find all MLRPs, as there may be more than one. It should be noted that the frequency counting in M 2 P is done by using a string matching algorithm, 10 since the frequency of a path P is equal to the number of appearances of P (i.e., of the sub-sequence corresponding to P) in S.…”

“…In [11,36] the retained characteristic of music is rhythm and the mapping used leads to rhythm strings. In [10] music melody and contour are mapped.Chen H. and Chen A.L.P. [12] focuses on properties such as pitch, duration and loudness, while [16,23,41] retain the melody of the music object.…”

“…Based on the well-studied text, image and video data IR, music IR can be performed using text (metadata) [2], pieces of structured or unstructured music [6,16,23,41], humming [6,10,15,17,31] or even classic western musical notation [32] as queries.…”