Automatic Tonic Pitch Estimation in South Indian Classical Music using Frequency- ratio Method

Computing techniques are nowadays widely used for searching for musical information (MIR). In southern Indian classical music, tonic identification is critical for data-based computational melodic analysis. Tonic is the fundamental pitch that a performer chooses and uses as a reference throughout an Indian classical music performance. Among the seven basic notes (swaras) used in Indian art music, the swara Sa, located in the middle octave range of performance, is the tonic pitch. Frequently, the tonic must be identified in real time to judge the quality of the singer or instrumentalist. In the proposed method, the attributes obtained from the pitch histogram using the frequency ratio approach are used to estimate the tonic. For tonic identification, a mathematical model is suggested based on the tonal histogram's properties. The performance is compared with existing frameworks, and the proposed technique outperforms the baseline methods. The proposed tonic identification system is also implemented using a Texas Instruments TMS320C6713 digital signal processor.