Notice of RetractionA Melody Matching and Retrieval Approach to Query by Humming in Music Retrieval System

Notice of Retraction

After careful and considered review of the content of this paper by a duly constituted expert committee, this paper has been found to be in violation of IEEE´s Publication Principles.

We hereby retract the content of this paper. Reasonable effort should be made to remove all past references to this paper.

The presenting author of this paper has the option to appeal this decision by contacting TPII@ieee.org.

In this paper, a method for representing medoly including pitch and note´s duration is defined,and we propose an integrated algorithm as well for evaluating melody similarity in query by humming system. In order to get a proper matching result, the two factors pitch and note´s duration are considered in our algorithm. Firstly,two-stage Hash index structure is provided for melody feature database;Secondly, Edit Distance algorithm is improved during the medoly calculation ;Thirdly, note´ duration is merged by analysing the executive path in Edit Distance algorithm and then rhythm similarity is calculated making use of vector angle cosin;Finally, the whole similarity is caculated combining the data of pitch with rhythm. The experiment shows that : this algorithm not only reduces system response time ,but performs better compared with traditional algorithm in top3 hit rate and error-tolerance.