Voice acting might not be viable in SmileBASIC. You can only record 1 second samples. Theoretically, you could combine a whole bunch of samples to create one large sound clip, but I feel like it could get pretty messy. I'd love to see someone give it a shot though.