Statistical modeling and retrieval of polyphonic music. Unal, E., Georgiou, P. G., Narayanan, S. S., & Chew, E. In 2007 IEEE 9Th International Workshop on Multimedia Signal Processing, MMSP 2007 - Proceedings, pages 405–409, Crete, 2007. doi abstract bibtex AbstractIn this article, we propose a solution to the problem of query by example for polyphonic music audio.We first present a generic mid-level representation for audio queries. Unlike previous efforts in the literature, the proposed representation is not dependent on the different spectral characteristics of different musical instruments and the accurate location of note onsets and offsets. This is achieved by first mapping the short term frequency spectrum of consecutive audio frames to the musical space (The Spiral Array) and defining a tonal identity with respect to center of effect that is generated by the spectral weights of the musical notes. We then use the resulting single dimensional text representations of the audio to create n-gram statistical sequence models to track the tonal characteristics and the behavior of the pieces. After performing appropriate smoothing, we build a collection of melodic n-gram models for testing. Using perplexity-based scoring, we test the likelihood of a sequence of lexical chords (an audio query) given each model in the database collection. Initial results show that, some variations of the input piece appears in the top 5 results 81pct of the time for whole melody inputs within a 500 polyphonic melody database. We also tested the retrieval engine for small audio clips. Using 25s segments, variations of the input piece are among the top 5 results 75pct of the time.
@InProceedings{ unal.ea2007-statistical,
author = {Unal, Erdem and Georgiou, Panayiotis G. and Narayanan,
Shrikanth S. and Chew, Elaine},
year = {2007},
title = {Statistical modeling and retrieval of polyphonic music},
abstract = {AbstractIn this article, we propose a solution to the
problem of query by example for polyphonic music audio.We
first present a generic mid-level representation for audio
queries. Unlike previous efforts in the literature, the
proposed representation is not dependent on the different
spectral characteristics of different musical instruments
and the accurate location of note onsets and offsets. This
is achieved by first mapping the short term frequency
spectrum of consecutive audio frames to the musical space
(The Spiral Array) and defining a tonal identity with
respect to center of effect that is generated by the
spectral weights of the musical notes. We then use the
resulting single dimensional text representations of the
audio to create n-gram statistical sequence models to
track the tonal characteristics and the behavior of the
pieces. After performing appropriate smoothing, we build a
collection of melodic n-gram models for testing. Using
perplexity-based scoring, we test the likelihood of a
sequence of lexical chords (an audio query) given each
model in the database collection. Initial results show
that, some variations of the input piece appears in the
top 5 results 81pct of the time for whole melody inputs
within a 500 polyphonic melody database. We also tested
the retrieval engine for small audio clips. Using 25s
segments, variations of the input piece are among the top
5 results 75pct of the time.},
address = {Crete},
booktitle = {2007 IEEE 9Th International Workshop on Multimedia Signal
Processing, MMSP 2007 - Proceedings},
doi = {10.1109/MMSP.2007.4412902},
isbn = {1424412749},
keywords = {computer and music},
mendeley-tags= {computer and music},
number = {November},
pages = {405--409}
}
Downloads: 0
{"_id":"JDnoQYS2amRf2vX3q","bibbaseid":"unal-georgiou-narayanan-chew-statisticalmodelingandretrievalofpolyphonicmusic-2007","authorIDs":[],"author_short":["Unal, E.","Georgiou, P. G.","Narayanan, S. S.","Chew, E."],"bibdata":{"bibtype":"inproceedings","type":"inproceedings","author":[{"propositions":[],"lastnames":["Unal"],"firstnames":["Erdem"],"suffixes":[]},{"propositions":[],"lastnames":["Georgiou"],"firstnames":["Panayiotis","G."],"suffixes":[]},{"propositions":[],"lastnames":["Narayanan"],"firstnames":["Shrikanth","S."],"suffixes":[]},{"propositions":[],"lastnames":["Chew"],"firstnames":["Elaine"],"suffixes":[]}],"year":"2007","title":"Statistical modeling and retrieval of polyphonic music","abstract":"AbstractIn this article, we propose a solution to the problem of query by example for polyphonic music audio.We first present a generic mid-level representation for audio queries. Unlike previous efforts in the literature, the proposed representation is not dependent on the different spectral characteristics of different musical instruments and the accurate location of note onsets and offsets. This is achieved by first mapping the short term frequency spectrum of consecutive audio frames to the musical space (The Spiral Array) and defining a tonal identity with respect to center of effect that is generated by the spectral weights of the musical notes. We then use the resulting single dimensional text representations of the audio to create n-gram statistical sequence models to track the tonal characteristics and the behavior of the pieces. After performing appropriate smoothing, we build a collection of melodic n-gram models for testing. Using perplexity-based scoring, we test the likelihood of a sequence of lexical chords (an audio query) given each model in the database collection. Initial results show that, some variations of the input piece appears in the top 5 results 81pct of the time for whole melody inputs within a 500 polyphonic melody database. We also tested the retrieval engine for small audio clips. Using 25s segments, variations of the input piece are among the top 5 results 75pct of the time.","address":"Crete","booktitle":"2007 IEEE 9Th International Workshop on Multimedia Signal Processing, MMSP 2007 - Proceedings","doi":"10.1109/MMSP.2007.4412902","isbn":"1424412749","keywords":"computer and music","mendeley-tags":"computer and music","number":"November","pages":"405–409","bibtex":"@InProceedings{ unal.ea2007-statistical,\n author = {Unal, Erdem and Georgiou, Panayiotis G. and Narayanan,\n Shrikanth S. and Chew, Elaine},\n year = {2007},\n title = {Statistical modeling and retrieval of polyphonic music},\n abstract = {AbstractIn this article, we propose a solution to the\n problem of query by example for polyphonic music audio.We\n first present a generic mid-level representation for audio\n queries. Unlike previous efforts in the literature, the\n proposed representation is not dependent on the different\n spectral characteristics of different musical instruments\n and the accurate location of note onsets and offsets. This\n is achieved by first mapping the short term frequency\n spectrum of consecutive audio frames to the musical space\n (The Spiral Array) and defining a tonal identity with\n respect to center of effect that is generated by the\n spectral weights of the musical notes. We then use the\n resulting single dimensional text representations of the\n audio to create n-gram statistical sequence models to\n track the tonal characteristics and the behavior of the\n pieces. After performing appropriate smoothing, we build a\n collection of melodic n-gram models for testing. Using\n perplexity-based scoring, we test the likelihood of a\n sequence of lexical chords (an audio query) given each\n model in the database collection. Initial results show\n that, some variations of the input piece appears in the\n top 5 results 81pct of the time for whole melody inputs\n within a 500 polyphonic melody database. We also tested\n the retrieval engine for small audio clips. Using 25s\n segments, variations of the input piece are among the top\n 5 results 75pct of the time.},\n address = {Crete},\n booktitle = {2007 IEEE 9Th International Workshop on Multimedia Signal\n Processing, MMSP 2007 - Proceedings},\n doi = {10.1109/MMSP.2007.4412902},\n isbn = {1424412749},\n keywords = {computer and music},\n mendeley-tags= {computer and music},\n number = {November},\n pages = {405--409}\n}\n\n","author_short":["Unal, E.","Georgiou, P. G.","Narayanan, S. S.","Chew, E."],"key":"unal.ea2007-statistical","id":"unal.ea2007-statistical","bibbaseid":"unal-georgiou-narayanan-chew-statisticalmodelingandretrievalofpolyphonicmusic-2007","role":"author","urls":{},"keyword":["computer and music"],"metadata":{"authorlinks":{}},"downloads":0},"bibtype":"inproceedings","biburl":"https://hmb.sampaio.me/bibliografia.bib.txt","creationDate":"2019-08-02T17:31:57.884Z","downloads":0,"keywords":["computer and music"],"search_terms":["statistical","modeling","retrieval","polyphonic","music","unal","georgiou","narayanan","chew"],"title":"Statistical modeling and retrieval of polyphonic music","year":2007,"dataSources":["n6MFY2CscQLDpJ7nT","RFLDZw5KyJdadDXDm"]}