issue with synthesized test sentence text

[superhg2012]

used the pre_trained model to run test sentence text, the output sample wavs sounds like noize， is it right?

[GeronimoKTT]

Hi man, can you tell me how you actually did used the pre_trained model to run text sentence text?
Cause I've been trying but it doesn't work, it keep show this error message :
"TypeError: must be encoded string without NULL bytes, not str"

Can you just tell me how you did? Thanks