Tonic-independent stroke transcription of the mridangam

Akshay Anantapadmanabhan, Juan P. Bello, Raghav Krishnan, Hema A. Murthy

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In this paper, we use a data-driven approach for the tonic-independent transcription of strokes of the mridangam, a South Indian hand drum. We obtain feature vectors that encode tonic-invariance by computing the magnitude spectrum of the constant-Q transform of the audio signal. Then we use Non-negative Matrix Factorization (NMF) to obtain a low-dimensional feature space where mridangam strokes are separable. We make the resulting feature sequence event-synchronous using short-term statistics of feature vectors between onsets, before classifying into a predefined set of stroke labels using Support Vector Machines (SVM). The proposed approach is both more accurate and flexible compared to that of tonic-specific approaches.

Original languageEnglish (US)
Title of host publication53rd AES International Conference 2014: Semantic Audio
PublisherAudio Engineering Society
Pages202-211
Number of pages10
ISBN (Print)9781632662842
StatePublished - 2014
Event53rd AES International Conference 2014: Semantic Audio - London, United Kingdom
Duration: Jan 26 2014Jan 29 2014

Other

Other53rd AES International Conference 2014: Semantic Audio
CountryUnited Kingdom
CityLondon
Period1/26/141/29/14

Fingerprint

Transcription
strokes
Invariance
Factorization
Support vector machines
Labels
audio signals
drums
Statistics
classifying
factorization
invariance
statistics
matrices

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Acoustics and Ultrasonics

Cite this

Anantapadmanabhan, A., Bello, J. P., Krishnan, R., & Murthy, H. A. (2014). Tonic-independent stroke transcription of the mridangam. In 53rd AES International Conference 2014: Semantic Audio (pp. 202-211). Audio Engineering Society.

Tonic-independent stroke transcription of the mridangam. / Anantapadmanabhan, Akshay; Bello, Juan P.; Krishnan, Raghav; Murthy, Hema A.

53rd AES International Conference 2014: Semantic Audio. Audio Engineering Society, 2014. p. 202-211.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Anantapadmanabhan, A, Bello, JP, Krishnan, R & Murthy, HA 2014, Tonic-independent stroke transcription of the mridangam. in 53rd AES International Conference 2014: Semantic Audio. Audio Engineering Society, pp. 202-211, 53rd AES International Conference 2014: Semantic Audio, London, United Kingdom, 1/26/14.
Anantapadmanabhan A, Bello JP, Krishnan R, Murthy HA. Tonic-independent stroke transcription of the mridangam. In 53rd AES International Conference 2014: Semantic Audio. Audio Engineering Society. 2014. p. 202-211
Anantapadmanabhan, Akshay ; Bello, Juan P. ; Krishnan, Raghav ; Murthy, Hema A. / Tonic-independent stroke transcription of the mridangam. 53rd AES International Conference 2014: Semantic Audio. Audio Engineering Society, 2014. pp. 202-211
@inproceedings{cc965c733aae4f4d80b99f144c9d8fb9,
title = "Tonic-independent stroke transcription of the mridangam",
abstract = "In this paper, we use a data-driven approach for the tonic-independent transcription of strokes of the mridangam, a South Indian hand drum. We obtain feature vectors that encode tonic-invariance by computing the magnitude spectrum of the constant-Q transform of the audio signal. Then we use Non-negative Matrix Factorization (NMF) to obtain a low-dimensional feature space where mridangam strokes are separable. We make the resulting feature sequence event-synchronous using short-term statistics of feature vectors between onsets, before classifying into a predefined set of stroke labels using Support Vector Machines (SVM). The proposed approach is both more accurate and flexible compared to that of tonic-specific approaches.",
author = "Akshay Anantapadmanabhan and Bello, {Juan P.} and Raghav Krishnan and Murthy, {Hema A.}",
year = "2014",
language = "English (US)",
isbn = "9781632662842",
pages = "202--211",
booktitle = "53rd AES International Conference 2014: Semantic Audio",
publisher = "Audio Engineering Society",
address = "United States",

}

TY - GEN

T1 - Tonic-independent stroke transcription of the mridangam

AU - Anantapadmanabhan, Akshay

AU - Bello, Juan P.

AU - Krishnan, Raghav

AU - Murthy, Hema A.

PY - 2014

Y1 - 2014

N2 - In this paper, we use a data-driven approach for the tonic-independent transcription of strokes of the mridangam, a South Indian hand drum. We obtain feature vectors that encode tonic-invariance by computing the magnitude spectrum of the constant-Q transform of the audio signal. Then we use Non-negative Matrix Factorization (NMF) to obtain a low-dimensional feature space where mridangam strokes are separable. We make the resulting feature sequence event-synchronous using short-term statistics of feature vectors between onsets, before classifying into a predefined set of stroke labels using Support Vector Machines (SVM). The proposed approach is both more accurate and flexible compared to that of tonic-specific approaches.

AB - In this paper, we use a data-driven approach for the tonic-independent transcription of strokes of the mridangam, a South Indian hand drum. We obtain feature vectors that encode tonic-invariance by computing the magnitude spectrum of the constant-Q transform of the audio signal. Then we use Non-negative Matrix Factorization (NMF) to obtain a low-dimensional feature space where mridangam strokes are separable. We make the resulting feature sequence event-synchronous using short-term statistics of feature vectors between onsets, before classifying into a predefined set of stroke labels using Support Vector Machines (SVM). The proposed approach is both more accurate and flexible compared to that of tonic-specific approaches.

UR - http://www.scopus.com/inward/record.url?scp=84904327014&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84904327014&partnerID=8YFLogxK

M3 - Conference contribution

SN - 9781632662842

SP - 202

EP - 211

BT - 53rd AES International Conference 2014: Semantic Audio

PB - Audio Engineering Society

ER -