Forecasting of COVID-19 daily report data has been one of the several challenges posed to the governments and health sectors on a global scale. Several deterministic and stochastic epidemiological models, including growth and compartmental models, have been proposed in the literature. These models assume that an epidemic would last over a short duration and the observed cases/deaths would attain a single peak. However, COVID-19 has extended over a longer duration than expected. Moreover, time-varying disease transmission rates due to government interventions such as business/school closures and openings have resulted in the observed data being multi-modal. This work proposes stochastic epidemiological models under a unified Bayesian framework augmented by a change-point detection mechanism to account for multiple peaks. We use the MCMC algorithms to sample the posterior distributions of model parameters for a fixed number of change-points. Additionally, we also develop a reversible jump MCMC algorithm to estimate the number of change-points and the resulting epidemiological parameters for allowing model parameter spaces to have a variable dimensionality.