Querying Large Collections of Music for Similarity

Matt Welsh; Nikita Borisov; Jason Hill; Robert von Behren

Abstract

We present a system capable of performing similarity queries against a large archive of digital music. Users are able to search for songs which "sound similar" to a given query song, thereby aiding the navigation and discovery of new music in such an archive. Our technique is based on reduction of the music data to a feature space of relatively small dimensionality (1248 feature dimensions per song); this is accomplished using a set of feature extractors which derive frequency, amplitude, and tempo data from the encoded music data. Queries are then performed using a <i>k</i>-nearest neighbor search in the feature space. Our system allows subsets of the feature space to be selected on a per-query basis. <p>We have integrated the music query engine into an online MP3 music archive consisting of over 7000 songs. We present an evaluation of our feature extraction and query results against this archive

Keywords

Computer scienceFeature (linguistics)Nearest neighbor searchSet (abstract data type)Information retrievalFeature vectorSimilarity (geometry)Space (punctuation)Digital audiok-nearest neighbors algorithmCurse of dimensionalityDimensionality reductionMusic information retrievalFeature extractionData miningArtificial intelligenceSpeech recognitionAudio signal

Related Publications

Similarity Search in High Dimensions via Hashing

Aristides Gionis , Piotr Indyk , Rajeev Motwani

The nearest- or near-neighbor query problems arise in a large variety of database applications, usually in the context of similarity searching. Of late, there has been increasin...

1999 3096 citations

Exploratory analysis and visualization of speech and music by locally linear embedding

Vaibhav Jain , L.K. Saul

Many problems in voice recognition and audio processing involve feature extraction from raw waveforms. The goal of feature extraction is to reduce the dimensionality of the audi...

2004 46 citations

Experiments on neural net recognition of spoken and written text

Dc Burr

The problems are discussed of the recognition of handprinted and spoken digits and the handprinted and spoken English alphabet. Four such experiments were conducted and the resu...

1988 IEEE Transactions on Acoustics Speech... 263 citations

Shape Matching and Object Recognition Using Low Distortion Correspondences

Alexander C. Berg , Tamara L. Berg , Jitendra Malik

We approach recognition in the framework of deformable shape matching, relying on a new algorithm for finding correspondences between feature points. This algorithm sets up corr...

2005 847 citations

Surface Learning with Applications to Lipreading

Christoph Bregler , Stephen M. Omohundro

Most connectionist research has focused on learning mappings from one space to another (eg. classification and regression). This paper introduces the more general task of learni...

1993 88 citations

Publication Info

Year: 2000
Type: article
Citations: 31
Access: Closed

External Links

Citation Metrics

OpenAlex

Cite This

APA Style

                            
                                    Matt Welsh, 
                                
                                    Nikita Borisov, 
                                
                                    Jason Hill
                                
                                et al.
                            
                            (2000). 
                            Querying Large Collections of Music for Similarity. 
                            
                            .