search

PABannier / bark.cpp

Suno AI's Bark model in C/C++ for fast text-to-speech
MIT License
635 stars 49 forks source link

MNT Add vocab file in data folder #78

Closed PABannier closed 11 months ago

vietanhdev commented 11 months ago

@PABannier That file needs to be in models/vocab.txt:

# Even if bark relies on GPT to encode text, it uses BertTokenizer (WordPiece)
    with open(dir_model / "vocab.txt", "r", encoding="utf-8") as fin:
        vocab = fin.readlines()
PABannier commented 11 months ago

@vietanhdev It depends on which dir_model you specify. You could choose models in which case, you should indeed place vocab.txt in models.