A Speech/Music Discriminator of Radio Recordings Based on Dynamic Programming and Bayesian Networks

Pikrakis, Aggelos; Γιαννακόπουλος, Θεόδωρος; Theodoridis, Sergios

doi:10.1109/tmm.2008.922870

Cited by 51 publications

(39 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…A related 2015 MIREX task used data from the British Library's music collections for a comparative study on the problem of speech/music segmentation. 13 This particular binary task has formed a research problem for over a decade Pikrakis et al (2008). Leaving the two-class scenario, SeFiRe, a downloadable tool for the segmentation and visualization of field recordings, segments and labels an audio file into five classes: speech, solo and choir singing, instrumental, and bell chiming.…”

Section: Existing Mir Tools: a Case Studymentioning

confidence: 99%

MIRchiving

Valk

Volk

Holzapfel

et al. 2017

Proceedings of the 4th International Workshop on Digital Libraries for Musicology

View full text Add to dashboard Cite

This study is a call for action for the music information retrieval (MIR) community to pay more attention to collaboration with digital music archives. The study, which resulted from an interdisciplinary workshop and subsequent discussion, matches the demand for MIR technologies from various archives with what is already supplied by the MIR community. We conclude that the expressed demands can only be served sustainably through closer collaborations. Whereas MIR systems are described in scientific publications, usable implementations are often absent. If there is a runnable system, user documentation is often sparse-posing a huge hurdle for archivists to employ it. This study sheds light on the current limitations and opportunities of MIR research in the context of music archives by means of examples, and highlights available tools. As a basic guideline for collaboration, we propose to interpret MIR research as part of a value chain. We identify the following benefits of collaboration between MIR researchers and music archives: new perspectives for content access in archives, more diverse evaluation data and methods, and a more application-oriented MIR research workflow.

show abstract

Section: Existing Mir Tools: a Case Studymentioning

confidence: 99%

MIRchiving

Valk

Volk

Holzapfel

et al. 2017

Proceedings of the 4th International Workshop on Digital Libraries for Musicology

View full text Add to dashboard Cite

show abstract

“…The signal amplitude measured in root mean square (RMS) and zero-crossing (ZC) are used in real time implementation of SMD [3]. The dynamic programming and Bayesian networks with the entropy of the normalized spectral energy are applied for SMD of radio recording in [4]. In [1], thirteen different audio features were used to train different types of multidimensional classifiers, including a Gaussian maximum, a posteriori (MAP) estimator and a nearest neighbor classifier.…”

Section: Introductionmentioning

confidence: 99%

Spectrogram based features selection using multiple kernel learning for speech/music discrimination

Nilufar

Ray

Molla

et al. 2012

2012 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

This paper presents a multiple kernel learning (MKL) approach to speech/music discrimination (SMD). The timefrequency representation (spectrogram) implemented by short-time Fourier transform (STFT) of audio segment is decomposed by wavelet packet transform into different subband levels. The subbands, which contain rich texture information, are used as features for this discrimination problem. MKL technique is used to select the optimal subbands to discriminate the audio signals. The proposed MKL based algorithm is applied for SMD of a standard dataset. The experimental results show that the proposed technique yields noticeable improvements in classification accuracy and tolerance toward different noise types compared to the existing methods.

show abstract

“…Previous work in the area of audio classification has focused mostly in audio event classification [1] and speech/music discrimination [2]- [4]. In this letter we investigate the problem of automatically classifying collections of audio files in three acoustic classes: speech, instrumental music and song (music with singing voice).…”

Section: Introductionmentioning

confidence: 99%

“…Among the different kind of features proposed for speech/music discrimination, it is worth mentioning the well-known Mel-Frequency Cepstral Coefficients (MFCC) [4], [6], Line Spectral Frequencies (LSF) [2], Zero-Crossing Rate (ZCR) and Frame Energy [4], [6] and more specific parameters such as Spectral Centroid, Spectral Flux, Spectral Rolloff [6] or Chroma-Vector based features [4].…”

Section: Introductionmentioning

confidence: 99%

“…The most common segment-based features are the mean and variance of a full segment [7], although higher order statistics such as kurtosis or skewness have been reported [6] or complex combinations of all these features (as, for example, the quotient between the maximum value of the feature in the segment and the mean [4]). The problem is that is not easy to determine a priori which kind of statistics of a certain parameter are more suitable for the classification task.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation