1
0
This commit is contained in:
mrq 2023-03-17 13:08:34 +00:00
parent 249c6019af
commit 96b7f9d2cc

View File

@ -1894,7 +1894,7 @@ def tokenize_text( text, stringed=True, skip_specials=False ):
if not tts:
tokenizer = VoiceBpeTokenizer(args.tokenizer_json if args.tokenizer_json else get_tokenizer_jsons()[0])
else:
tts.tokenizer
tokenizer = tts.tokenizer
encoded = tokenizer.encode(text)
decoded = tokenizer.tokenizer.decode(encoded, skip_special_tokens=skip_specials).split(" ")