Upload tokenizer

added_tokens.json

@@ -1,4 +1,4 @@
 {
-  "</s>": 51,
-  "<s>": 50
+  "</s>": 44,
+  "<s>": 43
 }

special_tokens_map.json

@@ -1,6 +1,6 @@
 {
   "bos_token": "<s>",
   "eos_token": "</s>",
-  "pad_token": "<pad>",
-  "unk_token": "<unk>"
+  "pad_token": "[PAD]",
+  "unk_token": "[UNK]"
 }

tokenizer_config.json

@@ -1,50 +1,49 @@
 {
   "added_tokens_decoder": {
-    "0": {
-      "content": "<pad>",
+    "41": {
+      "content": "[UNK]",
       "lstrip": true,
       "normalized": false,
       "rstrip": true,
       "single_word": false,
       "special": false
     },
-    "1": {
-      "content": "<s>",
+    "42": {
+      "content": "[PAD]",
       "lstrip": true,
       "normalized": false,
       "rstrip": true,
       "single_word": false,
       "special": false
     },
-    "2": {
-      "content": "</s>",
-      "lstrip": true,
+    "43": {
+      "content": "<s>",
+      "lstrip": false,
       "normalized": false,
-      "rstrip": true,
+      "rstrip": false,
       "single_word": false,
-      "special": false
+      "special": true
     },
-    "3": {
-      "content": "<unk>",
-      "lstrip": true,
+    "44": {
+      "content": "</s>",
+      "lstrip": false,
       "normalized": false,
-      "rstrip": true,
+      "rstrip": false,
       "single_word": false,
-      "special": false
+      "special": true
     }
   },
   "bos_token": "<s>",
   "clean_up_tokenization_spaces": false,
   "do_lower_case": false,
-  "do_normalize": true,
   "eos_token": "</s>",
   "extra_special_tokens": {},
   "model_max_length": 1000000000000000019884624838656,
-  "pad_token": "<pad>",
+  "pad_token": "[PAD]",
+  "processor_class": "Wav2Vec2Processor",
   "replace_word_delimiter_char": " ",
-  "return_attention_mask": false,
   "target_lang": null,
   "tokenizer_class": "Wav2Vec2CTCTokenizer",
-  "unk_token": "<unk>",
-  "word_delimiter_token": "/"
+  "unk_token": "[UNK]",
+  "word_delimiter_token": "|"
 }

vocab.json

@@ -1,58 +1,45 @@
 {
-  "$": 24,
-  "&": 8,
-  "'": 5,
-  "*": 20,
-  "/": 4,
-  "<": 9,
-  "</s>": 2,
-  "<pad>": 0,
-  "<s>": 1,
-  "<unk>": 3,
-  ">": 7,
-  "A": 11,
-  "D": 26,
-  "E": 29,
-  "F": 42,
-  "G": 55,
-  "H": 17,
-  "J": 53,
-  "K": 44,
-  "N": 43,
-  "P": 52,
-  "S": 25,
-  "T": 27,
-  "V": 54,
-  "Y": 40,
-  "Z": 28,
-  "_": 31,
-  "`": 50,
-  "a": 45,
-  "b": 12,
-  "d": 19,
-  "f": 32,
-  "g": 30,
-  "h": 38,
-  "i": 47,
-  "j": 16,
-  "k": 34,
-  "l": 35,
-  "m": 36,
-  "n": 37,
-  "o": 49,
-  "p": 13,
-  "q": 33,
-  "r": 21,
-  "s": 23,
-  "t": 14,
-  "u": 46,
-  "v": 15,
-  "w": 39,
-  "x": 18,
-  "y": 41,
-  "z": 22,
-  "{": 51,
-  "|": 6,
-  "}": 10,
-  "~": 48
+  "*": 1,
+  "+": 2,
+  "/": 3,
+  "2": 4,
+  "[": 5,
+  "[PAD]": 42,
+  "[UNK]": 41,
+  "]": 6,
+  "|": 0,
+  "~": 7,
+  "ء": 8,
+  "ؤ": 9,
+  "ئ": 10,
+  "ا": 11,
+  "ب": 12,
+  "ة": 13,
+  "ت": 14,
+  "ث": 15,
+  "ج": 16,
+  "ح": 17,
+  "خ": 18,
+  "د": 19,
+  "ذ": 20,
+  "ر": 21,
+  "ز": 22,
+  "س": 23,
+  "ش": 24,
+  "ص": 25,
+  "ض": 26,
+  "ط": 27,
+  "ظ": 28,
+  "ع": 29,
+  "غ": 30,
+  "ف": 31,
+  "ق": 32,
+  "ك": 33,
+  "ل": 34,
+  "م": 35,
+  "ن": 36,
+  "ه": 37,
+  "و": 38,
+  "ى": 39,
+  "ي": 40
 }