(Are you doing this with isotopically-labeled aniline? Otherwise I can't see how it's worth the trouble.)

If you use CH2Cl2 to extract, keep your aqueous phase volume low and filter your organic phase through silanized paper (e.g. Whatman 1PS) before letting it stand over drying agent. Also, CH2Cl2 can react with amines, so a quick literature search on the reaction with aniline may give you an idea of safe working temperatures. In any event, use only as much heat and vacuum as you need to evaporate solvent.

There's an alternate method in case solvent extraction gives poor recovery: Starting with (PhNH3+)2 SO42- solid, suspend in methanol and add a minimum amount water dropwise to dissolve. Add stoichiometric NaOH or KOH, either as a very concentrated water solution or as solution in methanol. Na2SO4 or K2SO4 will form a fine precipitate which is hard to remove - for the first pass, centrifugation is useful, but if you have to use filter paper then do a final pass through a glass fiber filter. Rotovap the solvent while monitoring both bath temp and vacuum to minimize evaporation of aniline. Drying can be done by redissolving the residual liquid in a small volume of CH2Cl2 and allowing to stand over anhydrous Na2SO4 or MgSO4 and re-evaporating, or you can use absolute ethanol and initially distill off solvent at atmospheric pressure, azeotroping water; avoid strong vacuum and keep bath heat low when removing the last of the solvent.