I'd like the new mobile vision text recognition API to scan for numbers or letters only. Does the API allow for this currently? Right now I do a lot of post processing to clean up errors like "b" is really a "6", or
If not now is there a plan for an API in the future?