Query by Singing/humming System Based on the Combination of Dtw Distances for Mirex 2012
نویسندگان
چکیده
This extended abstract descries KETI’s submission to the query-by-singing/humming (QbSH) task of MIREX 2012. Our QbSH system is based on dynamic time warping (DTW) and frame-based pitch sequence. Our system reduces false alarm to combine the distances of multiple DTW processes. To improve the performance, asymmetric DTW with boundary condition, compensation, and distances insensitive to the error are investigated.
منابع مشابه
Mirex2008: Query by Humming/singing System
This extended abstract describes my submission to the QBSH (Query by Singing/Humming) task of MIREX (Music Information Retrieval Evaluation eXchange) 2008. The system takes advantage of note-based and frame-based matching methods to improve the accuracy of the Query by Singing/Humming system. First, Earth Mover’s Distance (EMD), which is note-based and much faster, is adopted to eliminate most ...
متن کاملMultiple Classifiers-based Query by Singing/humming System
This extended abstract describes the method for MIREX 2011 task “Query by Singing/Humming”. The method extracts the pitch values from a database, and then the extracted data are normalized by zero elimination, meanshifting, median filtering, average filtering, and min-max scaling [1-4]. For matching, we used multiple classifiers. From those, a number of distances are calculated and combined bas...
متن کاملAn Improved Query by Singing/Humming System Using Melody and Lyrics Information
This paper proposes an improved query by singing/humming (QBSH) system using both melody and lyrics information for achieving better performance. Singing/humming discrimination (SHD) is first performed to distinguish singing from humming queries. For a humming query, we apply a pitch-only melody recognition method that has been used for QBSH task at MIREX with rank-1 performance. For a singing ...
متن کاملQBSH System for MIREX
This paper describes HCCL lab’s submission to the Queryby-Singing/Humming(QBSH) task of Music Information Retrieval eXchange(MIREX) 2006. As we do not participate in the second sub-task, this paper will only deal with the Known-Item Retrieval sub-task. In the submitted system, we apply a novel algorithm called Recursive Alignment(RA) to compute the similarity score between query and candidates....
متن کاملQuery by Humming Using Locality Sensitive Hashing for Mirex 2008
We submit a method for the MIREX 2008 task “Query by Singing/Humming” based on locality sensitive hashing (LSH). The method constructs an index of melodic fragments by extracting pitch vectors from a database of melodies. In retrieval, the method automatically transcribes a sung query into notes and then extracts pitch vectors similarly to the index construction. For each query pitch vector, th...
متن کامل