Adding new scores
Browse files
NbAiLab_NPSC_16K_mp3_test_eval_results.txt
CHANGED
@@ -1,2 +1,2 @@
|
|
1 |
-
WER: 0.
|
2 |
-
CER: 0.
|
|
|
1 |
+
WER: 0.09829976810645934
|
2 |
+
CER: 0.031738041959668346
|
eval.py
CHANGED
@@ -65,6 +65,7 @@ def normalize_text(text: str) -> str:
|
|
65 |
text = re.sub('<ee>', 'eee', text)
|
66 |
text = re.sub('<qq>', 'qqq', text)
|
67 |
text = re.sub('<mm>', 'mmm', text)
|
|
|
68 |
text = re.sub('[<>]', '', text)
|
69 |
|
70 |
# # 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:
|
|
80 |
def main(args):
|
81 |
# load dataset
|
82 |
dataset = load_dataset(args.dataset, args.config, split=args.split, use_auth_token=True)
|
|
|
83 |
|
84 |
# for testing: only process the first two examples as a test
|
85 |
# dataset = dataset.select(range(10))
|
|
|
65 |
text = re.sub('<ee>', 'eee', text)
|
66 |
text = re.sub('<qq>', 'qqq', text)
|
67 |
text = re.sub('<mm>', 'mmm', text)
|
68 |
+
text = re.sub('<inaudible>', 'xxx', text)
|
69 |
text = re.sub('[<>]', '', text)
|
70 |
|
71 |
# # In addition, we can normalize the target text, e.g. removing new lines characters etc...
|
|
|
81 |
def main(args):
|
82 |
# load dataset
|
83 |
dataset = load_dataset(args.dataset, args.config, split=args.split, use_auth_token=True)
|
84 |
+
#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))
|
85 |
|
86 |
# for testing: only process the first two examples as a test
|
87 |
# dataset = dataset.select(range(10))
|
log_NbAiLab_NPSC_16K_mp3_test_predictions.txt
CHANGED
The diff for this file is too large to render.
See raw diff
|
|
log_NbAiLab_NPSC_16K_mp3_test_targets.txt
CHANGED
The diff for this file is too large to render.
See raw diff
|
|