A Generative Model for Rhythms

A Generative Model for Rhythms. Paiement, J., Grandvalet, Y., Bengio, S., & Eck, D. In NIPS Workshop on Brain, Music and Cognition, 2007.

Paper abstract bibtex

Modeling music involves capturing long-term dependencies in time series, which has proved very difficult to achieve with traditional statistical methods. The same problem occurs when only considering rhythms. In this paper, we introduce a generative model for rhythms based on the distributions of distances between subsequences. A specific implementation of the model when considering Hamming distances over a simple rhythm representation is described. The proposed model consistently outperforms a standard Hidden Markov Model in terms of conditional prediction accuracy on two different music databases.

@inproceedings{paiement:2007:nips,
  author = {J.-F. Paiement and Y. Grandvalet and S. Bengio and D. Eck},
  title = {A Generative Model for Rhythms},
  booktitle = {NIPS Workshop on Brain, Music and Cognition},
  year = 2007,
  url = {publications/ps/paiement_2007_nips.ps.gz},
  pdf = {publications/pdf/paiement_2007_nips.pdf},
  djvu = {publications/djvu/paiement_2007_nips.djvu},
  original = {2007/rhythm_pred_nips},
  topics = {graphical_models},
  abstract = {Modeling music involves capturing long-term dependencies in time series, which has proved very difficult to achieve with traditional statistical methods. The same problem occurs when only considering rhythms. In this paper, we introduce a generative model for rhythms based on the distributions of distances between subsequences.  A specific implementation of the model when considering Hamming distances over a simple rhythm representation is described. The proposed model consistently outperforms a standard Hidden Markov Model in terms of conditional prediction accuracy on two different music databases.},
  categorie = {C},
}

Downloads: 0

{"_id":"gBmpSDkKzBnyfvkcz","bibbaseid":"paiement-grandvalet-bengio-eck-agenerativemodelforrhythms-2007","authorIDs":[],"author_short":["Paiement, J.","Grandvalet, Y.","Bengio, S.","Eck, D."],"bibdata":{"bibtype":"inproceedings","type":"inproceedings","author":[{"firstnames":["J.-F."],"propositions":[],"lastnames":["Paiement"],"suffixes":[]},{"firstnames":["Y."],"propositions":[],"lastnames":["Grandvalet"],"suffixes":[]},{"firstnames":["S."],"propositions":[],"lastnames":["Bengio"],"suffixes":[]},{"firstnames":["D."],"propositions":[],"lastnames":["Eck"],"suffixes":[]}],"title":"A Generative Model for Rhythms","booktitle":"NIPS Workshop on Brain, Music and Cognition","year":"2007","url":"publications/ps/paiement_2007_nips.ps.gz","pdf":"publications/pdf/paiement_2007_nips.pdf","djvu":"publications/djvu/paiement_2007_nips.djvu","original":"2007/rhythm_pred_nips","topics":"graphical_models","abstract":"Modeling music involves capturing long-term dependencies in time series, which has proved very difficult to achieve with traditional statistical methods. The same problem occurs when only considering rhythms. In this paper, we introduce a generative model for rhythms based on the distributions of distances between subsequences. A specific implementation of the model when considering Hamming distances over a simple rhythm representation is described. The proposed model consistently outperforms a standard Hidden Markov Model in terms of conditional prediction accuracy on two different music databases.","categorie":"C","bibtex":"@inproceedings{paiement:2007:nips,\n author = {J.-F. Paiement and Y. Grandvalet and S. Bengio and D. Eck},\n title = {A Generative Model for Rhythms},\n booktitle = {NIPS Workshop on Brain, Music and Cognition},\n year = 2007,\n url = {publications/ps/paiement_2007_nips.ps.gz},\n pdf = {publications/pdf/paiement_2007_nips.pdf},\n djvu = {publications/djvu/paiement_2007_nips.djvu},\n original = {2007/rhythm_pred_nips},\n topics = {graphical_models},\n abstract = {Modeling music involves capturing long-term dependencies in time series, which has proved very difficult to achieve with traditional statistical methods. The same problem occurs when only considering rhythms. In this paper, we introduce a generative model for rhythms based on the distributions of distances between subsequences. A specific implementation of the model when considering Hamming distances over a simple rhythm representation is described. The proposed model consistently outperforms a standard Hidden Markov Model in terms of conditional prediction accuracy on two different music databases.},\n categorie = {C},\n}\n\n","author_short":["Paiement, J.","Grandvalet, Y.","Bengio, S.","Eck, D."],"key":"paiement:2007:nips","id":"paiement:2007:nips","bibbaseid":"paiement-grandvalet-bengio-eck-agenerativemodelforrhythms-2007","role":"author","urls":{"Paper":"http://bengio.abracadoudou.com/publications/ps/paiement_2007_nips.ps.gz"},"downloads":0},"bibtype":"inproceedings","biburl":"http://bengio.abracadoudou.com/samy.bib","creationDate":"2020-03-18T03:43:27.361Z","downloads":0,"keywords":[],"search_terms":["generative","model","rhythms","paiement","grandvalet","bengio","eck"],"title":"A Generative Model for Rhythms","year":2007,"dataSources":["9NCW2CDr4M3s5DvNX"]}