With a rapidly increasing number of photometric time series becoming available for O- and B-type stars, especially since the launch of TESS, it has become clear that stochastic low-frequency (SLF) variability is a common type of variability found in massive stars. This type of variability is seen as an increase in power towards low frequencies, as characteristic of red noise, and has been suggested to be caused by internal gravity waves, subsurface convection zones, radiatively driven stellar winds, or a combination thereof. Here we present the analysis and characterization of the SLF variability of 50 O- and B-type stars observed by TESS, of which 49 are found in six different Cygnus OB associations and one is in the Small Magellanic Cloud. We adopt two different methods for characterizing the signal. For the first we fit a Lorentzian-like function to the power density spectrum of the stars. Our second model-independent method uses the root-mean-square of the photometric variability and the cumulative integrated power density to characterize the variability. We compare our results to predictions from 2D and 3D simulations of internal gravity waves, subsurface convection, and stellar winds, finding good agreement between the observed characteristic frequencies of our sample and the predictions from sub-surface convection.