That sounds kinda fucked up to do to a child. I wouldn't do it with mine. Okay, operant conditioning seems to work on children only and that only temporary. The Default Mode Network (DMN) associated with self-referential thought is not innate in children, but developed through late childhood, adolescence, and up to around 25 when the brain reaches maturation. It seems to me that operant conditioning works for as long as self-referential thought and sense of self aren't yet developed. I'd fear the formation of cathexis to the parents. Externalization of their speech to the parents who serve as a source of validation might then stunt natural development of their own identity, sense of self, and self-reliance. Classical conditioning does not seem to have therapeutic intervention as I saw, but rather serve as a framework to understand its development. Classical conditioning seems like the natural progression after the operant. I'll leave the operant conditioning for my dog who I know won't develop his own person and higher-order thinking. Did you know hummingbirds may develop stutters? Physical trauma or God knows what. They typically recover from it after 4 months though, even if their speech doesn't return entirely the same as it used to be.