Fix inability to load vocab.json on converting the 16B model due to encoding of the file not being set

[swanserquack]

Today I was trying to convert the 16B model using the python script but kept runing into the error
UnicodeDecodeError: 'charmap' codec can't decode byte 0x81 in position 926: character maps to <undefined>

having a look into it, the error seems to occur as the encoding of the file is not set to UTF-8 when it is opened by the script. Tested change on my personal machine and it converted the model fine without any errors

[ravenscroftj]

Nice spot thank you - the vocab.json files for all of the models should be utf-8 encoded I think so I'm happy to accept this change. I definitely need to add some automated regression tests for PRs in future :)