wav2vec2-large-tedlium / tokenizer_config.json
sanchit-gandhi's picture
add tokenizer
e5eaaa7
raw
history blame
338 Bytes
{
"bos_token": "<s>",
"do_lower_case": false,
"eos_token": "</s>",
"name_or_path": "sanchit-gandhi/wav2vec2_ctc_tedlium_tokenizer",
"pad_token": "<pad>",
"replace_word_delimiter_char": " ",
"special_tokens_map_file": null,
"tokenizer_class": "Wav2Vec2CTCTokenizer",
"unk_token": "<unk>",
"word_delimiter_token": "|"
}