nathanhunt
/

wav2vec2-large-mms-1b-aka-colab

Automatic Speech Recognition

Inference Endpoints

Model card Files Files and versions Community

nathanhunt commited on Sep 1, 2023

Commit

ee36210

•

1 Parent(s): 949de36

Upload tokenizer

Files changed (3) hide show

added_tokens.json +3 -2
special_tokens_map.json +7 -0
vocab.json +3 -5

added_tokens.json CHANGED Viewed

@@ -1,4 +1,5 @@
 {
-  "</s>": 29,
-  "<s>": 28
 }

 {
+  "</s>": 26,
+  "<s>": 25,
+  "[PAD]": 27
 }

special_tokens_map.json CHANGED Viewed

@@ -13,6 +13,13 @@
       "normalized": true,
       "rstrip": false,
       "single_word": false
     }
   ],
   "bos_token": "<s>",

       "normalized": true,
       "rstrip": false,
       "single_word": false
+    },
+    {
+      "content": "[PAD]",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false
     }
   ],
   "bos_token": "<s>",

vocab.json CHANGED Viewed

@@ -1127,8 +1127,7 @@
     "ꞌ": 7
   },
   "aka": {
-    " ": 0,
-    "[PAD]": 27,
     "[UNK]": 26,
     "a": 1,
     "b": 2,
@@ -1150,11 +1149,10 @@
     "u": 18,
     "w": 19,
     "y": 20,
     "á": 21,
     "ɔ": 22,
-    "ɛ": 23,
-    "’": 24,
-    "•": 25
   },
   "akb": {
     "'": 25,

     "ꞌ": 7
   },
   "aka": {
+    "[PAD]": 26,
     "[UNK]": 26,
     "a": 1,
     "b": 2,
     "u": 18,
     "w": 19,
     "y": 20,
+    "|": 0,
     "á": 21,
     "ɔ": 22,
+    "ɛ": 23
   },
   "akb": {
     "'": 25,