AI-SUPPORTED PROCESSING OF HANDWRITTEN TRANSCRIPTIONS FOR HUNGARIAN FOLK SONGS IN A DIGITAL ENVIRONMENT

My research focuses on creating an AI-supported Digital Research Environment (DRE) that helps analysing and systematizing folk music tunes with the help of latest information theory database management results. The study may be ex- tended to entire source material accumulated by researchers so far, thus inte- grating Hungarian ethnomusicology results last hundred years. In this way, new dimensions structural analysis open up a large amount can processed already exceeds limits human musical memory. Previous computerized experiments in Hungary have inadequate- ly defined role artificial intelligence. our case, digital en- vironment is subject does not work independently, because researcher’s scientifically abstract thinking, preferences, recognition characteristic melodic elements cannot yet replaced computer data processing. Crucial goal precisely define musi- cal Thus attitude rejecting software support 1 institute previously belonged Academy Science, currently it belongs ELKH (Eötvös Lóránd Network). 2 List publications: MTMT. Scientific Bibliography. URL: https:// m2.mtmt.hu/gui2/?type=authors&mode=browse&sel=10063399 (Access: 23.10.2022). https://doi.org/10.33398/2523-4846-2022-18-1-65-82 66 change favour actually using framework. For first time Hungar- ian history, detailed documented environ- ment created, integrating useful, relevant tools. We map out entry problems standard format suitable for mass input analysis. If possible, we will replace widely used op- tional scalable broader range parametrization search options, their free combination allows us scientific models. With DRE, validity previous classification more specified processing as well unreported melodies process type significantly accelerated. most significant debate has been dataset speci- fication analyses. I am convinced only similarly tune-data-elements compared, one critical tasks determine data’s density. As step, conversion manuscript needs solved. International mainly led printed music, some which project, but many developments are also needed. Keywords: (DRE), Optical Music Recognition (OMR), Musical Manuscripts, Folk Songs, classification, ethnomusicology, archives, folklore database.