wav2vec2-mms-1b-all-swc-kat5 / tokenizer_config.json
Ussen's picture
Upload tokenizer
e65f1b5
raw
history blame
355 Bytes
{
"bos_token": "<s>",
"clean_up_tokenization_spaces": true,
"do_lower_case": false,
"eos_token": "</s>",
"model_max_length": 1000000000000000019884624838656,
"pad_token": "[PAD]",
"replace_word_delimiter_char": " ",
"target_lang": "swc",
"tokenizer_class": "Wav2Vec2CTCTokenizer",
"unk_token": "[UNK]",
"word_delimiter_token": "|"
}