Upload tokenizer

README.md

@@ -1,11 +1,11 @@
 ---
+base_model: facebook/wav2vec2-xls-r-300m
 library_name: transformers
 license: apache-2.0
-base_model: facebook/wav2vec2-xls-r-300m
-tags:
-- generated_from_trainer
 metrics:
 - wer
+tags:
+- generated_from_trainer
 model-index:
 - name: wav2vec2-large-xls-r-300m-tsovatush-demo-colab-bryn-hauk-9-21
   results: []

added_tokens.json

@@ -1,4 +1,4 @@
 {
-  "</s>": 49,
-  "<s>": 48
+  "</s>": 47,
+  "<s>": 46
 }

tokenizer_config.json

@@ -1,6 +1,6 @@
 {
   "added_tokens_decoder": {
-    "46": {
+    "44": {
       "content": "[UNK]",
       "lstrip": true,
       "normalized": false,
@@ -8,7 +8,7 @@
       "single_word": false,
       "special": false
     },
-    "47": {
+    "45": {
       "content": "[PAD]",
       "lstrip": true,
       "normalized": false,
@@ -16,7 +16,7 @@
       "single_word": false,
       "special": false
     },
-    "48": {
+    "46": {
       "content": "<s>",
       "lstrip": false,
       "normalized": false,
@@ -24,7 +24,7 @@
       "single_word": false,
       "special": true
     },
-    "49": {
+    "47": {
       "content": "</s>",
       "lstrip": false,
       "normalized": false,

vocab.json

@@ -1,7 +1,7 @@
 {
   "'": 1,
-  "[PAD]": 47,
-  "[UNK]": 46,
+  "[PAD]": 45,
+  "[UNK]": 44,
   "a": 2,
   "b": 3,
   "c": 4,
@@ -33,18 +33,16 @@
   "ħ": 29,
   "ŏ": 30,
   "š": 31,
-  "ŭ": 32,
-  "ž": 33,
-  "ǧ": 34,
-  "ǰ": 35,
-  "ɣ": 36,
-  "ɬ": 37,
-  "ʒ": 38,
-  "ʔ": 39,
-  "ʕ": 40,
-  "ʡ": 41,
-  "ʲ": 42,
-  "ʷ": 43,
-  "ː": 44,
-  "ⁿ": 45
+  "ž": 32,
+  "ǯ": 33,
+  "ǰ": 34,
+  "ɣ": 35,
+  "ɬ": 36,
+  "ʒ": 37,
+  "ʔ": 38,
+  "ʕ": 39,
+  "ʡ": 40,
+  "ʷ": 41,
+  "ː": 42,
+  "ⁿ": 43
 }