Bauyrjan commited on
Commit
2d83161
1 Parent(s): d797097

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
+ {"в": 0, "ч": 1, "ы": 3, "ө": 4, "г": 5, "щ": 6, "у": 7, "л": 8, "ш": 9, "т": 10, "н": 11, "м": 12, "о": 13, "–": 14, "ӊ": 15, "х": 16, "ц": 17, "й": 18, "с": 19, "ё": 20, "ф": 21, "ю": 22, "э": 23, "ң": 24, "ь": 25, "я": 26, "и": 27, "⅛": 28, "д": 29, "б": 30, "п": 31, "ж": 32, "к": 33, "—": 34, "е": 35, "ъ": 36, "ү": 37, "а": 38, "р": 39, "з": 40, "|": 2, "[UNK]": 41, "[PAD]": 42}