Model trained for longer

README.md

@@ -113,9 +113,9 @@ result = test_dataset.map(evaluate, batched=True, batch_size=8)
 print("WER: {:2f}".format(100 * wer.compute(predictions=result["pred_strings"], references=result["sentence"])))
 ```
 
-**Test Result**: 23.33 %
+**Test Result**: 22.27 %
 
 
 ## Training
 
-The [NST Swedish Dictation](https://www.nb.no/sprakbanken/en/resource-catalogue/oai-nb-no-sbr-17/) was used for training. The [Fairseq](https://github.com/fairseq) scripts were used.
+The [NST Swedish Dictation](https://www.nb.no/sprakbanken/en/resource-catalogue/oai-nb-no-sbr-17/) was used for training. The [Fairseq](https://github.com/fairseq) scripts were used.

config.json

@@ -64,5 +64,5 @@
   "num_hidden_layers": 24,
   "pad_token_id": 0,
   "transformers_version": "4.4.0.dev0",
-  "vocab_size": 35
+  "vocab_size": 36
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dc24d4db3a895f9c8042cee59749fee054d594f135c35ad58d4e2cc9a5578a17
-size 1262071447
+oid sha256:135a6af4d856277695a7cd99c4791961d8cc67347c2ba10bf3b870deb368a7a9
+size 1262075543

vocab.json

@@ -1 +1 @@
-{"<pad>": 0, "<s>": 1, "</s>": 2, "<unk>": 3, "|": 4, "T": 5, "E": 6, "A": 7, "R": 8, "N": 9, "S": 10, "I": 11, "L": 12, "D": 13, "O": 14, "M": 15, "G": 16, "K": 17, "F": 18, "U": 19, "V": 20, "H": 21, "Ä": 22, "Å": 23, "P": 24, "Ö": 25, "B": 26, "J": 27, "C": 28, "Y": 29, "X": 30, "É": 31, "Z": 32, "W": 33 }
+{"<pad>": 0, "<s>": 1, "</s>": 2, "<unk>": 3, "|": 4, "T": 5, "E": 6, "A": 7, "R": 8, "N": 9, "S": 10, "I": 11, "L": 12, "D": 13, "O": 14, "M": 15, "G": 16, "K": 17, "U": 18, "V": 19, "H": 20, "F": 21, "Ä": 22, "Å": 23, "P": 24, "Ö": 25, "B": 26, "C": 27, "J": 28, "Y": 29, "X": 30, "É": 31, "-": 32, "Z": 33, "W": 34, "Q": 35 }