Conventional and periodic N-grams in the transcription of drum sequences

Paulus, Jouni; Klapuri, Anssi

doi:10.1109/icme.2003.1221722

Cited by 16 publications

(27 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…As a consequence, drum loop signals or drum tracks often exhibit a temporal structure. Two concurrent studies have exploited such a structure by means of a sequence model, or "language model" by analogy with large vocabulary speech recognition systems ( [Paulus and Klapuri, 2003] for drum sequences transcription, or [Gillet and Richard, 2003] for the transcription of tabla signals).…”

Section: Introductionmentioning

confidence: 99%

Drum Loops Retrieval from Spoken Queries

Gillet

Richard

2005

J Intell Inf Syst

View full text Add to dashboard Cite

Recent efforts in audio indexing and music information retrieval mostly focus on melody. If this is appropriate for polyphonic music signals, specific approaches are needed for systems dealing with percussive audio signals such as those produced by drums, tabla or djembé. In this article, we present a complete system allowing the management of a drum loops database. Queries in this database are formulated with spoken onomatopoeias -short meaningless words imitating the different sounds of the drumkit. The transcription task necessary to index the database is performed using Hidden Markov Models (HMM) and Support Vector Machines (SVM) and achieves a 86.4% correct recognition rate. The syllables of spoken queries are recognized and a relevant statistical model allows the comparison and alignment of the query with the rythmic sequences stored in the database, in order to provide a set of the most relevant drum loops.

show abstract

Section: Introductionmentioning

confidence: 99%

Drum Loops Retrieval from Spoken Queries

Gillet

Richard

2005

J Intell Inf Syst

View full text Add to dashboard Cite

show abstract

“…Herrera et al [11] compared conventional classifiers in experiments on identifying individual drum sounds. To transcribe drum sounds in drums-only audio signals, the use of N-grams [14], probabilistic models [15], and HMM&SVM [6] have been proposed. To identify drum sounds extracted from polyphonic audio signals, Sandvolt et al [17] proposed a feature-model adaptation method that is robust to the distortion of features since feature distortion caused by other sounds is a major problem.…”

Section: Related Workmentioning

confidence: 99%

INTER:D: a drum sound equalizer for controlling volume and timbre of drums

Yoshii¹,

Goto²,

Okuno³

2005

2nd European Workshop on the Integration of Knowledge, Semantics and Digital Media Technology (EWIMT 2005)

View full text Add to dashboard Cite

A drum sound equalizer, called INTER:D, is described that enables a listener to control the volume and timbre of bass and snare drum sounds in commercial compact-disc recordings. Although the characteristics of the drum sounds are often closely related to the impression made by a musical piece, conventional graphic equalizers cannot adjust their characteristics because they have volume sliders only for the frequency bands. INTER:D provides drum-specific volume sliders that directly control the frequency components of drum sounds, enabling a listener to intuitively cut or boost the volume of each drum. In addition, INTER:D enables a listener to replace the original timbre of each drum with another timbre selected from a dropdown list. These interactive functions are achieved using an automatic music content analysis system based on low-level audio signal processing. This system can estimate the power spectrogram of each drum sound and detect the onset times in a musical piece. Subjective experiments showed that INTER:D can add a new dimension to the way users experience music.

show abstract

“…As noted in [9], this may lead to the occurrence of very unlikely combinations in the transcription. The co-occurrence information is modelled implicitly with the combination modelling.…”

Section: Using the Modelsmentioning

confidence: 93%

“…However, in practice a small portion of these combinations contribute the majority of all combination occurrences [9,3]. This observation can be used to choose a subset of combinations to be modelled.…”

Section: Combination Modellingmentioning

confidence: 99%

Acoustic Modelling of Drum Sounds with Hidden Markov Models for Music Transcription

Paulus

2006 IEEE International Conference on Acoustics Speed and Signal Processing Proceedings

View full text Add to dashboard Cite

This paper describes two methods for applying hidden Markov models (HMMs) to acoustic modelling of drum sound events for polyphonic music transcription. The proposed methods are instrumentwise binary modelling and modelling of instrument combinations. In the first, each target instrument is modelled with a "sound" model and all target instruments share a "silence" model. Each instrument is transcribed independently from the others. In the latter method, different instrument combinations are modelled, and an additional "silence" model is created. The proposed methods are evaluated with simulations with acoustic data, and compared with two reference methods. Simulations show that combination modelling performs better than instrument-wise modelling.

show abstract

Conventional and periodic N-grams in the transcription of drum sequences

Cited by 16 publications

References 9 publications

Drum Loops Retrieval from Spoken Queries

Drum Loops Retrieval from Spoken Queries

INTER:D: a drum sound equalizer for controlling volume and timbre of drums

Acoustic Modelling of Drum Sounds with Hidden Markov Models for Music Transcription

Contact Info

Product

Resources

About