DrishtiSharma commited on
Commit
3554927
1 Parent(s): 08f8b7b

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
+ {"ᱚ": 1, "ᱛ": 2, "ᱜ": 3, "ᱝ": 4, "ᱞ": 5, "ᱟ": 6, "ᱠ": 7, "ᱡ": 8, "ᱢ": 9, "ᱣ": 10, "ᱤ": 11, "ᱥ": 12, "ᱦ": 13, "ᱧ": 14, "ᱨ": 15, "ᱩ": 16, "ᱪ": 17, "ᱫ": 18, "ᱬ": 19, "ᱭ": 20, "ᱮ": 21, "ᱯ": 22, "ᱰ": 23, "ᱱ": 24, "ᱲ": 25, "ᱴ": 26, "ᱵ": 27, "ᱶ": 28, "ᱷ": 29, "ᱸ": 30, "ᱹ": 31, "ᱼ": 32, "ᱽ": 33, "᱾": 34, "|": 0, "[UNK]": 35, "[PAD]": 36}