Multimodal Aspects of Music Retrieval: Audio, Song Lyrics – and Beyond?

Mayer, Rudolf; Rauber, Andreas

doi:10.1007/978-3-642-11674-2_15

Cited by 11 publications

(10 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For instance, Porter and Neuringer [347] test whether pigeons trained to discriminate between music by J. S. Bach and Stravinsky are able to discriminate between music by composers contemporary with J. S. Bach (Buxtehude and Scarlatti) and Stravinsky (Carter and Piston). Another extension of Classify is Scale [9,13,14,19,45,48,49,53,62,68,83,94,132,144,199,225,226,238,261,275,280,303,320,336,337,[339][340][341]390] [425] use Classify in comparing rhythmic features (statistics of an autocorrelation of wavelet decomposition) and timbral features (spectral centroid, rolloff, etc.). On the other hand, Yoon et al [459] explore two different feature selection approaches using Classify, but do not discuss or list the selected features.…”

Section: 7mentioning

confidence: 99%

A Survey of Evaluation in Music Genre Recognition

Sturm

2014

Adaptive Multimedia Retrieval: Semantics, Context, and Adaptation

View full text Add to dashboard Cite

Abstract. Much work is focused upon music genre recognition (MGR) from audio recordings, symbolic data, and other modalities. While reviews have been written of some of this work before, no survey has been made of the approaches to evaluating approaches to MGR. This paper compiles a bibliography of work in MGR, and analyzes three aspects of evaluation: experimental designs, datasets, and figures of merit.

show abstract

Section: 7mentioning

confidence: 99%

A Survey of Evaluation in Music Genre Recognition

Sturm

2014

Adaptive Multimedia Retrieval: Semantics, Context, and Adaptation

View full text Add to dashboard Cite

show abstract

“…Studies that discuss about the relationships between lyrical expressions and affections cannot be founded. Song selection methods using both acoustic and lyric have been proposed [3,4]. However, there are few studies focusing on "lyrical expressions" and affections.…”

Section: Related Workmentioning

confidence: 99%

Detecting Attractive Phrases from Musical Lyrics Focusing on Typical Expressions on Lyrics

Yamanishi

Kagita

Nishihara

et al. 2015

IJAE

View full text Add to dashboard Cite

Lyrics influence human affection toward music as same as the acoustics. All phrases in a lyric are not remembered; there are some "attractive phrases" in lyrics. This paper proposes a method to detect attractive phrases from musical lyrics focusing on typical expressions on lyrics that are uncommon in natural document. Through the interview on impression for lyrics, "uniqueness of co-occurred terms" and "repetition" were founded as typical expressions that significantly attracted human. Therefore, these expressions were modeled as mathematical features. The proposed method detected attractive phrases using support vector machine with the modeled features. The results of the model evaluation experiments showed the 69% accuracy and 86% precision. The detected results were compared and discussed with the key sentences detected by using the existing summarization methods. As the result of the comparison, it was concluded that the proposed method detected attractive phrases more accurate than the existing summarization methods.

show abstract

“…Thus, thresholding may need to be taken on the 2 histogram determined by using the Ostu's method [29], and a pixel is claimed to have motion if either 1 or 2 is above the corresponding thresholds. Figure 2.17a shows the framework of the HS algorithm with a 3*3 mask and resulting 2*9 data matrices [44]. In Fig.…”

Section: Motion Analysismentioning

confidence: 99%

“…2.17d after Kalman filtering. [44] There are two other classic optical flow algorithms, namely, the Lucas-Kanade method [45] and the Brox method [36]. The Lucas-Kanade method assumes that the flow is essentially constant in a local neighborhood of the pixel.…”

Section: Motion Analysismentioning

confidence: 99%

“…To track progress in text reading, a Robust Reading Competition was held at ICDAR 2011 dealing with two challenges. The first challenge deals with born-digital images (web, e-mail) whereas the second challenge deals with scene text recognition [44]. In ICDAR 2011, the dataset is extended with more images containing a variety of outdoor and indoor scenes.…”

Section: Scene Text Datasetsmentioning

confidence: 99%

See 1 more Smart Citation