In the 1980s, the Nanyang Municipal Bureau of Culture and the Nanyang Mass Art
Center in Henan Province jointly collected and compiled The Anthology of Nanyang Folk
Songs, which documented over 450 folk songs representative of the musical traditions of
Central China. The original audio recordings, once preserved by the Nanyang Mass Art
Center, have been completely lost due to historical and environmental factors, leaving
only textual and notational records. This study explores how these “silent archives” can
be digitally reconstructed and recontextualized in the contemporary era. By applying AI
voice modeling and quantitative analysis of melodic contour, modal structure, and scale
organization, combined with dialect-based voice synthesis, metadata design, and
geographic visualization, the research seeks to recreate the sonic image of Nanyang’s
folk tradition. The project aims to provide a new approach to digitally restoring lost folk
songs through artificial intelligence, offering possibilities for the preservation,
revitalization, and reproduction of traditional music in the digital age.