I think the biggest hurdle would be the overall flow. yes, you can cut words from other media, but you can't paste them into sentances unless all the words are in the same tone, and have the same inflection. Aso, it would be difficult to extract any background noise that may be playing as the voice actor is talking. The most I've ever done is make a music video to a TV show using the same voice actor from another series, but because I had music playing it was easier to drown out the background noise that came with the voice.