Title | ||
---|---|---|
An approach to transcription of varnams in Carnatic music using hidden Markov models. |
Abstract | ||
---|---|---|
In this paper, we present an approach to transcription of varnams in Carnatic music using hidden Markov models (HMMs). We draw upon domain-specific information in that we build raga-specific HMMs and use rhythmic cycle (tala) information in the initialization of the HMMs and in the post-processing of the output. The ground truth is obtained from standard notations of varnams. Transcription performance on varnams in eight ragas is presented. Since the extant notation for Carnatic music is incomplete when it comes to gamakas (melodic ornamentations), the results are interpreted accordingly. Significantly, they point to the possibility of a rich transcription for Carnatic music, which is an unsolved problem in the automatic processing of Indian music. |
Year | Venue | Field |
---|---|---|
2017 | National Conference on Communications NCC | Melody,Histogram,Notation,Computer science,Speech recognition,Music of India,Ground truth,Initialization,Hidden Markov model,Rhythm |
DocType | Citations | PageRank |
Conference | 0 | 0.34 |
References | Authors | |
8 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
P. V. Krishnaraj Sekhar | 1 | 0 | 0.34 |
Venkata Subramanian Viraraghavan | 2 | 1 | 4.40 |
Sridharan Sankaran | 3 | 0 | 0.34 |
Hema A. Murthy | 4 | 603 | 79.54 |