I really don't see the need for building 10 submarines annually. The bottleneck will always be crew training. With currently just 9 boats in service there's no way to reliably add more than 2-3 boats per year. Any additional submarines above this number are a waste of resources. As the size of the fleet increases 4-5 might become possible by the end of the decade but we should remember that new SSNs will also be tasked with more extensive missions as the fleet grows and subs become more capable.
I think training is less than a constraint than you think.
Instead of just looking at existing SSNs in service, I would add the fleet of 60 SSKs into consideration, as there is a lot of overlap in terms of skills and experience. That's a total of 70 submarines to draw upon.
From a training perspective, I would split it into technical personnel who maintain the submarine and then command personnel who use the submarine.
The technical personnel can conduct much of the training onshore. And you can just have conversion training for existing experienced submariners.
For the command (tactical) staff, you can promote the best Captains, Executive Officers, etc from the existing fleet. The only way they can really train on a new submarine class is to actually get real-life experience with its capabilities across the oceans.
If China were to add say 6 submarines per year, they would only have to promote the best 10% from the existing fleet of 70 submarines and do conversion training, before they're ready to go.