Recording and recognizing similar recurring sounds

I would like to record a child babbling to identify recurring similar sounds.

Duration and volume would not be relevant. Identification can take time, does not need to be prompt.

I understand Arduino is not powerful enough to manage complex sound information, but I read about different problems having different solutions (whose theories I usually do not understand).

If there is one, I wonder if anyone could kindly suggest an available project I can refer to.

Thank you in advance.