Upload processor

preprocessor_config.json

@@ -0,0 +1,10 @@
+{
+  "do_normalize": true,
+  "feature_extractor_type": "Wav2Vec2FeatureExtractor",
+  "feature_size": 1,
+  "padding_side": "right",
+  "padding_value": 0.0,
+  "processor_class": "Wav2Vec2Processor",
+  "return_attention_mask": true,
+  "sampling_rate": 16000
+}

special_tokens_map.json

@@ -1,6 +1,30 @@
 {
-  "bos_token": "<s>",
-  "eos_token": "</s>",
-  "pad_token": "[PAD]",
-  "unk_token": "[UNK]"
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "[PAD]",
+    "lstrip": true,
+    "normalized": false,
+    "rstrip": true,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "[UNK]",
+    "lstrip": true,
+    "normalized": false,
+    "rstrip": true,
+    "single_word": false
+  }
 }

tokenizer_config.json

@@ -39,6 +39,7 @@
   "eos_token": "</s>",
   "model_max_length": 1000000000000000019884624838656,
   "pad_token": "[PAD]",
+  "processor_class": "Wav2Vec2Processor",
   "replace_word_delimiter_char": " ",
   "target_lang": null,
   "tokenizer_class": "Wav2Vec2CTCTokenizer",

vocab.json

@@ -1,53 +1,53 @@
 {
   "[PAD]": 50,
   "[UNK]": 49,
-  "|": 33,
+  "|": 28,
   "ㄱ": 1,
-  "ㄲ": 20,
-  "ㄳ": 7,
-  "ㄴ": 17,
-  "ㄵ": 18,
-  "ㄶ": 10,
-  "ㄷ": 0,
-  "ㄸ": 8,
-  "ㄹ": 14,
-  "ㄺ": 44,
-  "ㄻ": 5,
-  "ㄼ": 21,
-  "ㅀ": 2,
-  "ㅁ": 23,
-  "ㅂ": 28,
-  "ㅃ": 9,
-  "ㅄ": 35,
-  "ㅅ": 36,
-  "ㅆ": 47,
+  "ㄲ": 33,
+  "ㄳ": 15,
+  "ㄴ": 10,
+  "ㄵ": 38,
+  "ㄶ": 16,
+  "ㄷ": 12,
+  "ㄸ": 7,
+  "ㄹ": 39,
+  "ㄺ": 46,
+  "ㄻ": 23,
+  "ㄼ": 4,
+  "ㅀ": 26,
+  "ㅁ": 27,
+  "ㅂ": 40,
+  "ㅃ": 31,
+  "ㅄ": 17,
+  "ㅅ": 32,
+  "ㅆ": 30,
   "ㅇ": 6,
-  "ㅈ": 24,
-  "ㅉ": 43,
-  "ㅊ": 41,
-  "ㅋ": 12,
-  "ㅌ": 25,
-  "ㅍ": 11,
-  "ㅎ": 13,
-  "ㅏ": 19,
-  "ㅐ": 22,
-  "ㅑ": 32,
-  "ㅒ": 45,
-  "ㅓ": 16,
-  "ㅔ": 40,
-  "ㅕ": 4,
-  "ㅖ": 42,
-  "ㅗ": 46,
-  "ㅘ": 3,
-  "ㅙ": 34,
-  "ㅚ": 26,
-  "ㅛ": 37,
-  "ㅜ": 38,
-  "ㅝ": 39,
-  "ㅞ": 27,
-  "ㅟ": 15,
-  "ㅠ": 29,
-  "ㅡ": 30,
-  "ㅢ": 48,
-  "ㅣ": 31
+  "ㅈ": 13,
+  "ㅉ": 19,
+  "ㅊ": 18,
+  "ㅋ": 9,
+  "ㅌ": 24,
+  "ㅍ": 35,
+  "ㅎ": 47,
+  "ㅏ": 0,
+  "ㅐ": 36,
+  "ㅑ": 22,
+  "ㅒ": 44,
+  "ㅓ": 41,
+  "ㅔ": 34,
+  "ㅕ": 20,
+  "ㅖ": 48,
+  "ㅗ": 21,
+  "ㅘ": 14,
+  "ㅙ": 2,
+  "ㅚ": 29,
+  "ㅛ": 25,
+  "ㅜ": 37,
+  "ㅝ": 3,
+  "ㅞ": 42,
+  "ㅟ": 45,
+  "ㅠ": 11,
+  "ㅡ": 43,
+  "ㅢ": 8,
+  "ㅣ": 5
 }