Exportar Publicação

Raposo, F., Ribeiro, R. & de Matos, D. M. (2016). Using generic summarization to improve music information retrieval tasks. IEEE/ACM Transactions on Audio, Speech and Language Processing. 24 (6), 1119-1128

F. A. Raposo et al.,  "Using generic summarization to improve music information retrieval tasks", in IEEE/ACM Transactions on Audio, Speech and Language Processing, vol. 24, no. 6, pp. 1119-1128, 2016

@article{raposo2016_1714724538729,
	author = "Raposo, F. and Ribeiro, R. and de Matos, D. M.",
	title = "Using generic summarization to improve music information retrieval tasks",
	journal = "IEEE/ACM Transactions on Audio, Speech and Language Processing",
	year = "2016",
	volume = "24",
	number = "6",
	doi = "10.1109/TASLP.2016.2541299",
	pages = "1119-1128",
	url = "http://ieeexplore.ieee.org/xpl/tocresult.jsp?isnumber=7463555"
}

TY  - JOUR
TI  - Using generic summarization to improve music information retrieval tasks
T2  - IEEE/ACM Transactions on Audio, Speech and Language Processing
VL  - 24
IS  - 6
AU  - Raposo, F.
AU  - Ribeiro, R.
AU  - de Matos, D. M.
PY  - 2016
SP  - 1119-1128
SN  - 2329-9290
DO  - 10.1109/TASLP.2016.2541299
UR  - http://ieeexplore.ieee.org/xpl/tocresult.jsp?isnumber=7463555
AB  - In order to satisfy processing time constraints, many music information retrieval (MIR) tasks process only a segment of the whole music signal. This may lead to decreasing performance, as the most important information for the tasks may not be in the processed segments. We leverage generic summarization algorithms, previously applied to text and speech, to summarize items in music datasets. These algorithms build summaries (both concise and diverse), by selecting appropriate segments from the input signal, also making them good candidates to summarize music. We evaluate the summarization process on binary and multiclass music genre classification tasks, by comparing the accuracy when using summarized datasets against the accuracy when using human-oriented summaries, continuous segments (the traditional method used for addressing the previously mentioned time constraints), and full songs of the original dataset. We show that GRASSHOPPER, LexRank, LSA, MMR, and a Support Sets-based centrality model improve classification performance when compared to selected baselines. We also show that summarized datasets lead to a classification performance whose difference is not statistically significant from using full songs. Furthermore, we make an argument stating the advantages of sharing summarized datasets for future MIR research.
ER  -