Recent ground-breaking work shows that stochastic volatility models driven by fractional Brownian motion with short memory provide better calibration of the volatility surface and more robust estimation of historical volatility. Based on the fractional nature of the volatility correlation, we choose two types of stochastic-local volatility built on the constant elasticity of variance model to calculate European option prices. Our analysis reveals how the associated implied volatility term structures are related to the elasticity factor and the Hurst exponent as well as the underlying value.