We consider a semi-Markov modulated market consisting of a riskless asset or bond, B, and a risky asset or stock, S, whose dynamics depend on a semi-Markov process x. Using the martingale characterization of semi-Markov processes, we note the incompleteness of semi-Markov modulated markets and find the minimal martingale measure. We price variance (Theorem 1) and volatility (Theorem 2) swaps for stochastic volatilities driven by the semi-Markov processes. We also discuss some extensions of the obtained results such as local semi-Markov volatility, Dupire formula for the local semi-Markov volatility and residual risk associated with the swap pricing.