wav2vec2-large-TIMIT-IPA / tokenizer_config.json
speech31's picture
add tokenizer
1a61a0c
raw
history blame contribute delete
236 Bytes
{
"bos_token": "<s>",
"do_lower_case": false,
"eos_token": "</s>",
"pad_token": "[PAD]",
"replace_word_delimiter_char": " ",
"tokenizer_class": "Wav2Vec2CTCTokenizer",
"unk_token": "[UNK]",
"word_delimiter_token": "|"
}