Volatile acids are usually associated with undesirable “sour” and “bitter” descriptors. The fermentation of high sugar grape juice/must using chilled, botrytised or late harvested grapes often leads to the production of higher amounts of volatile acidity, which adversely impacts the overall wine quality. This work aims to minimize the formation of volatile acidity, and further improve aroma complexity during high sugar fermentation. The potential application of indigenous Torulaspora delbruekii R12 was evaluated using late harvested Petit Manseng from east foot hill of the Helan Mountain. Petit Manseng juice with 300 g/L of sugar was filter sterilized prior to inoculation. Three inoculation strategies were used: 1) single fermentation with either pure R12 or indigenous Saccharomyces cerevisiae NX11424 at 1×106 cells/mL; 2) co-inoculation of R12 and NX11424 at the rate of 5:1, 20:1, and 50:1, among which the inoculum of R12 was 1×106 cells/mL; 3) 1×106 cells/mL of R12 was inoculated prior to the inoculation of NX11424 after 5 d at the same rate. Fermentation samples were collected every 48 h to measure the residual sugar and the formation of acetic acid using enzymatic analysis. Yeast viability was also determined via serial dilution and plating on WLN agar medium. Fermentation was terminated with the addition of 60 mg/L SO2 on the day of the 20th. Final ferments were centrifuged and stored for subsequent analysis on volatile compounds via head-space-solid phase micro extraction–gas chromatography coupled with mass spectrometry (HS-SPME-GC-MS). NX11424 monoculture fermentation was rapid, when utilizing 255 g/L of sugar in 20 d, whereas, the sugar consumption of R12 fermentations was relatively slower, almost halted at 20 d. Nevertheless, the fermentation power was much stronger in all the co-inoculated fermentations than that in both the R12 monoculture fermentations and the sequentially inoculated fermentations. Correspondingly, the viability of both yeast strains in each fermentation was inversely related to sugar consumption. In terms of acetic acid(the major component responsible for volatile acidity), NX11424 monoculture fermentation produced 1.2 g/L acetic acids, which fell just around the legal threshold. By contrast, there was a significant decrease in the amount of acetic acid for both co-inoculation with R12 at the ratio of 5:1, 20:1, and 50:1, and sequential inoculation. The reduction of acetic acid was in line with the increased proportion of R12 in the mixed inoculum, with the highest decrease being 48.9% at 50:1 co-fermentation, compared with the single fermentation with NX11424. Another noticeable effect was that significantly less abundant acetaldehyde related to oxidative descriptors appeared in wines produced with the combined R12 and NX11424. The reduction of this compound was up to 50% in the mixed culture fermentation, compared with the NX11424 monoculture fermentation. Further, the impact of R12 on aroma profiles of wine was evaluated, where 42 volatile compounds were detected by HS-SPME-GC-MS in Petit Manseng wines. It was found that the application of R12 was significantly correlated with the decrease of higher alcohols up to 30 mg/L, compared with the S. cerevisiae control. Significant differences were also observed in the concentration of esters. Specifically, the presence of R12 increased the level of isobutyl aetate, isoamyl acetate, hexyl acetate, and hexyl butyrate, whereas, remarkably reduced the production of phenethyl acetate, ethyl octanoate, ethyl decanoate, and ethyl dodecanoate. Lower concentrations of acetoin were also found in the wine samples involving R12. Additionally, a principal component analysis was utilized to clearly separate volatile compounds, where R12 inoculation strategies displayed a distinctive impact on wine aroma profile. In particular, the co-inoculation at the ratio of 20:1 behaved with the greatest potential to enhance both the varietal and the fermentative aromas of the wine. In this scenario, the amount of varietal volatile compounds was remarkably improved, such as terpenes, and C13 demethyl isoprene, whereas, a noticeable increase was also observed in the typical volatile compounds (eg., isoamyl acetate, ethyl butanoate, and phenyl ethanol)derived from fermentation. Therefore, the indigenous T. delbruekii R12 was expected to serve in conjunction with S.cerevisiae, thereby reducing acetic acid for better aroma quality during fermentation with high sugar in grape juice/must. The findings expand current knowledge on the solutions to efficiently minimizing volatile acidity during high sugar fermentations.