I'm not a logician, but I think the question (a good one) is not quite well-posed. As I understand, in mathematical logic, a statement being provable means one can deduce it from a set of axioms. If you have a set of axioms A sufficiently powerful for Godel's construction to go through, the resulting Godel sentence s is (by construction) not provable from A, but one can always make it provable by appending s to the axioms A.
For a statement to be true, on the other hand, has a technical meaning: it means that there is a model (in the sense of model theory, see https://en.wikipedia.org/wiki/Model_theory ) in which both the axioms A and the sentence s are true. Godel's completeness theorem (https://en.wikipedia.org/wiki/G%C3%B6del%27s_completeness_th...) asserts that s is provable from A if and only if s is true in every model in which A holds.
For a statement to be true, on the other hand, has a technical meaning: it means that there is a model (in the sense of model theory, see https://en.wikipedia.org/wiki/Model_theory ) in which both the axioms A and the sentence s are true. Godel's completeness theorem (https://en.wikipedia.org/wiki/G%C3%B6del%27s_completeness_th...) asserts that s is provable from A if and only if s is true in every model in which A holds.