The detection and quantification of wine yeast can be misleading due to under or overestimation of these microorganisms. Underestimation may be caused by variable growing rates of different microorganisms in culture media or the presence of viable but non-cultivable microorganisms. Overestimation may be caused by the lack of discrimination between live and dead microorganisms if quantitative PCR is used to quantify with DNA as the template. However, culture-independent methods that use dyes have been described to remove the DNA from dead cells and then quantify the live microorganisms. Two dyes have been studied in this paper: ethidium monoazide bromide (EMA) and propidium monoazide bromide (PMA). The technique was applied to grape must fermentation and ageing wines. Both dyes presented similar results on yeast monitoring. Membrane cell recovery was necessary when yeasts were originated from ethanol-containing media. When applied to grape must fermentation, differences of up to 1 log unit were seen between the QPCR estimation with or without the dye during the stationary phase. In ageing wines, good agreement was found between plating techniques and QPCR. Most of the viable cells were also culturable and no differences were observed with the methods, except for Zygosaccharomyces bailii and Dekkera bruxellensis where much higher counts were occasionally detected by QPCR. The presence of excess dead cells did not interfere with the quantification of live cells with either of the dyes.