Automatic Tonic Identification in Indian Art Music: Approaches and Evaluation

Abstract

The tonic is a fundamental concept in Indian art music. It is the base pitch, which an artist chooses in order to construct the melodies during a rāg(a) rendition, and all accompanying instruments are tuned using the tonic pitch. Consequently, tonic identification is a fundamental task for most computational analyses of Indian art music, such as intonation analysis, melodic motif analysis and rāg recognition. In this paper we review existing approaches for tonic identification in Indian art music and evaluate them on six diverse datasets for a thorough comparison and analysis. We study the performance of each method in different contexts such as the presence/absence of additional metadata, the quality of audio data, the duration of audio data, music tradition (Hindustani/Carnatic) and the gender of the singer (male/female). We show that the approaches that combine multi-pitch analysis with machine learning provide the best performance in most cases (90% identification accuracy on average), and are robust across the aforementioned contexts compared to the approaches based on expert knowledge. In addition, we also show that the performance of the latter can be improved when additional metadata is available to further constrain the problem. Finally, we present a detailed error analysis of each method, providing further insights into the advantages and limitations of the methods.

Mining Melodic Patterns in Large Audio Collections of Indian Art Music

“…Thus, our representation becomes independent of the tonic of the lead artist, which allows a meaningful comparison of melodies of two distinct recordings (even if sung by two different artists in different tonic pitches). The tonic of the lead artist is identified automatically using a classification-based multi-pitch approach [24]. We use the implementation of this method available in Essentia with the default set of parameters.…”

Morty

“…Considering the lack of open test datasets for makam recognition and the drawbacks of the available tonic identification datasets, we have gathered a test dataset of audio recordings with annotated makam and tonic, called the Ottoman-Turkish makam recognition dataset. 13 The dataset covers 20 commonly performed makams 14 and it is composed of 1000 audio recordings. Following our constraint in the problem definition, a single makam is performed in each recording (i.e.…”

“…Similarly, several studies on tonic identification use pitch distribution based methods [6,10]. More recently there has been an interest in culture specific methods for this task [2,14,22] that make use of heuristics and the musical characteristics of the studied tradition.…”

Culture-Aware Approaches to Modeling and Description of Intonation Using Multimodal Data

“…Srinivasamurthy et al [12] review several state-of-the-art approaches to rhythm description and evaluate them on Indian art music traditions and Makam music of Turkey. Gulati et al [13] evaluate various tonic identification approaches over Indian art music traditions using diverse datasets under different conditions of the data (E.g., duration of the input audio clip, presence/absence of metadata etc). Rao et al [14] propose methods to classify melodic patterns in Indian art music and evaluate them using retrieval experiments.…”