Real-Time Maqam Estimation Model in Max/MSP Configured for the Nāy

Automatic maqam estimation is consideredsignificant toward improving multimedia live music performances and automaticaccompaniment. This contribution proposed a real-time maqam estimation modeldeveloped in the visual programming language MAX/MSP and configured for thenāydukah. The model’s design stood on basic formulas of Arab music maqamat asexplained in theory and applied in practice. The model consisted of differentlayers of competition; the first was for the identification of the instanttonic of the melodic figure, and the second was for the recognition of itsidentifying E (E, E half-flat and E flat). Those two competitions were used toestimate the maqam in real-time. Then, accumulated estimation results wereused to estimate the maqam in longer durations; five-second and full duration.The model was evaluated using professionally performed nāy improvisations.Results reflected a success in estimating all the studied maqamat when the fullimprovisation was considered. In addition, results were very good for real-time and five-second estimation where average estimation confidence was 75.98%and 80.04%, respectively.