We analyze the properties of the indirect inference estimator when the observed series are contaminated by measurement error. We show that the indirect inference estimates are asymptotically biased when the nuisance parameters of the measurement error distribution are neglected in the indirect estimation. We propose to solve this inconsistency by jointly estimating the nuisance and the structural parameters. The range of applicability of this methodology is supported by theoretical results based on several examples for both discrete and continuous-time models. Indirect inference is used to estimate the parameters of stochastic volatility models with auxiliary specifications based on realized volatility measures. Monte Carlo simulations show the bias reduction of the indirect estimates obtained when the microstructure noise is explicitly modeled. Finally, an empirical application illustrates the relevance of a realistic specification of the microstructure noise distribution to match the features of the observed log-returns at high frequencies.