Siyong commited on
Commit
da96f7b
·
1 Parent(s): bf8c0f2

add tokenizer

Browse files
Files changed (3) hide show
  1. special_tokens_map.json +1 -0
  2. tokenizer_config.json +1 -0
  3. vocab.json +1 -0
special_tokens_map.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"bos_token": "<s>", "eos_token": "</s>", "unk_token": "[UNK]", "pad_token": "[PAD]"}
tokenizer_config.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"unk_token": "[UNK]", "bos_token": "<s>", "eos_token": "</s>", "pad_token": "[PAD]", "do_lower_case": false, "word_delimiter_token": "|", "replace_word_delimiter_char": " ", "tokenizer_class": "Wav2Vec2CTCTokenizer"}
vocab.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"k": 0, "n": 1, "f": 3, "g": 4, "d": 5, "o": 6, "l": 7, "h": 8, "b": 9, "e": 10, "p": 11, "u": 12, "m": 13, "v": 14, "a": 15, "y": 16, "r": 17, "q": 18, "s": 19, "x": 20, "z": 21, "i": 22, "c": 23, "j": 24, "'": 25, "w": 26, "t": 27, "[UNK]": 28, "[PAD]": 29, "|": 2}