Adding new scores

NbAiLab_NPSC_16K_mp3_test_eval_results.txt

@@ -1,2 +1,2 @@
-WER: 0.11892896728078826
-CER: 0.03728901318225468
+WER: 0.09829976810645934
+CER: 0.031738041959668346

eval.py

@@ -65,6 +65,7 @@ def normalize_text(text: str) -> str:
     text = re.sub('<ee>', 'eee', text)
     text = re.sub('<qq>', 'qqq', text)
     text = re.sub('<mm>', 'mmm', text)
+    text = re.sub('<inaudible>', 'xxx', text)
     text = re.sub('[<>]', '', text)
 
     # # In addition, we can normalize the target text, e.g. removing new lines characters etc...
@@ -80,6 +81,7 @@ def normalize_text(text: str) -> str:
 def main(args):
     # load dataset
     dataset = load_dataset(args.dataset, args.config, split=args.split, use_auth_token=True)
+    #dataset = load_dataset(args.dataset, args.config, split=args.split, use_auth_token=True).filter(lambda entry: re.search("nb-nn", entry["sentence_language_code"], flags=re.IGNORECASE))
 
     # for testing: only process the first two examples as a test
     # dataset = dataset.select(range(10))