Abstract: This study utilized Xinjiang Bosten Lake grass carp as the raw material, and the ultrasound-assisted vacuum marination method was employed to enhance the deep processing utilization rate and flavor quality of grass carp. Single-factor experiments and fuzzy mathematics sensory evaluation combined with response surface analysis were used to optimize the marination process conditions. Headspace-solid phase microextraction-gas chromatography-mass spectrometry (HS-SPME-GC-MS) was employed to analyze the effects of different drying methods, including cold air and hot air, on the volatile flavor components of grass carp blocks. The results showed that the fuzzy mathematics sensory score reached 87.6 with an ultrasound power of 200 W, a vacuum pressure of 0.06 MPa, an ultrasound time of 30 min, and a salt addition of 1.5%, indicating that the dried fish had good quality under these optimal conditions. 126 volatile components were detected in the cold-air-dried and hot-air-dried samples by HS-SPME-GC-MS. The variable importance in the projection (VIP) of the partial least squares-discriminant analysis (PLS-DA) model, combined with heatmap analysis, showed significant differences in fish meat flavor under different drying methods. Eight differential flavor compounds with VIP>1 that exhibited higher levels under cold air drying were identified, mainly including aromatic hydrocarbons, alkanes, and pinenes. In contrast, 15 compounds were identified in the hot air drying method, including esters, ketones, alcohols, and terpenes. There were various types of flavor compounds in hot air-dried grass carp, among which the content of esters and alcohols was significantly higher than that in the cold air drying method, adding fruity and fatty aromas to grass carp samples, and the overall flavor characteristics are richer and more intense. This study provides a reference for optimizing the drying process, flavor analysis, quality improvement, and product upgrading of local specialty dried grass carp.