radhakri119 commited on
Commit
381cc8f
1 Parent(s): 7fbf907

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": "|", "tokenizer_class": "Wav2Vec2CTCTokenizer"}
vocab.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"k": 0, "q": 2, "r": 3, "h": 4, "a": 5, "s": 6, "l": 7, "j": 8, "o": 9, "z": 10, "d": 11, "f": 12, "m": 13, "b": 14, "p": 15, "'": 16, "w": 17, "t": 18, "e": 19, "u": 20, "i": 21, "n": 22, "y": 23, "g": 24, "v": 25, "c": 26, "x": 27, "|": 1, "[UNK]": 28, "[PAD]": 29}