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.

Authors: Anantapadmanabhan, Akshay; Bello, Juan; Krishnan, Raghav; Murthy, Hema
Affiliations: Indian Institute of Technology Madras, Chennai Tamil Madu, India; New York University, New York, NY, USA(See document for exact affiliation information.)
AES Conference: 53rd International Conference: Semantic Audio (January 2014)
Paper Number: P2-2
Publication Date: January 27, 2014
Subject: Automatic Music Transcription

Click to purchase paper as a non-member or you can login as an AES member to see more options.

No AES members have commented on this paper yet.

Subscribe to this discussion

To be notified of new comments on this paper you can subscribe to this RSS feed. Forum users should login to see additional options.

Start a discussion!

If you would like to start a discussion about this paper and are an AES member then you can login here:
Username:
Password:

If you are not yet an AES member and have something important to say about this paper then we urge you to join the AES today and make your voice heard. You can join online today by clicking here.