Upload tokenizer

README.md

@@ -1,11 +1,11 @@
 ---
+base_model: facebook/w2v-bert-2.0
+datasets:
+- common_voice_16_0
 library_name: transformers
 license: mit
-base_model: facebook/w2v-bert-2.0
 tags:
 - generated_from_trainer
-datasets:
-- common_voice_16_0
 model-index:
 - name: w2v-bert-2.0-mongolian-colab-CV16.0
   results: []

added_tokens.json

@@ -1,4 +1,4 @@
 {
-  "</s>": 38,
-  "<s>": 37
+  "</s>": 36,
+  "<s>": 35
 }

tokenizer_config.json

@@ -1,6 +1,6 @@
 {
   "added_tokens_decoder": {
-    "35": {
+    "33": {
       "content": "[UNK]",
       "lstrip": true,
       "normalized": false,
@@ -8,7 +8,7 @@
       "single_word": false,
       "special": false
     },
-    "36": {
+    "34": {
       "content": "[PAD]",
       "lstrip": true,
       "normalized": false,
@@ -16,7 +16,7 @@
       "single_word": false,
       "special": false
     },
-    "37": {
+    "35": {
       "content": "<s>",
       "lstrip": false,
       "normalized": false,
@@ -24,7 +24,7 @@
       "single_word": false,
       "special": true
     },
-    "38": {
+    "36": {
       "content": "</s>",
       "lstrip": false,
       "normalized": false,
@@ -39,7 +39,6 @@
   "eos_token": "</s>",
   "model_max_length": 1000000000000000019884624838656,
   "pad_token": "[PAD]",
-  "processor_class": "Wav2Vec2BertProcessor",
   "replace_word_delimiter_char": " ",
   "target_lang": null,
   "tokenizer_class": "Wav2Vec2CTCTokenizer",

vocab.json

@@ -1,6 +1,6 @@
 {
-  "[PAD]": 36,
-  "[UNK]": 35,
+  "[PAD]": 34,
+  "[UNK]": 33,
   "|": 0,
   "а": 1,
   "б": 2,
@@ -22,18 +22,16 @@
   "с": 18,
   "т": 19,
   "у": 20,
-  "ф": 21,
-  "х": 22,
-  "ц": 23,
-  "ч": 24,
-  "ш": 25,
-  "ъ": 26,
-  "ы": 27,
-  "ь": 28,
-  "э": 29,
-  "ю": 30,
-  "я": 31,
-  "ё": 32,
-  "ү": 33,
-  "ө": 34
+  "х": 21,
+  "ц": 22,
+  "ч": 23,
+  "ш": 24,
+  "ы": 25,
+  "ь": 26,
+  "э": 27,
+  "ю": 28,
+  "я": 29,
+  "ё": 30,
+  "ү": 31,
+  "ө": 32
 }