The documentation does not indicate any way to constrain the individual normal distributions to have identical standard deviations. You can force the distributions to be independent, but that is not helpful in this case.
You can observe the estimated standard deviations of each distribution in the mixture after fitting this model.