Update README.md

README.md

@@ -28,4 +28,22 @@ language:
 size_categories:
 - 10K<n<100K
 ---
-common voice, google fleurs, JSUTv1.1, JAS_v2 (joujiboi/japanese-anime-speech-v2) processed for whisper. Not shuffled or normalized. 50% anime speech and 50% other. Other corpus's are fully represented. 
+common voice, google fleurs, JSUTv1.1, JAS_v2 (joujiboi/japanese-anime-speech-v2) processed for whisper. Not shuffled or normalized. 50% anime speech and 50% other. Other corpus's are fully represented.
+
+if needed:
+
+import neologdn
+import MeCab
+import re
+from transformers.models.whisper.english_normalizer import BasicTextNormalizer
+wakati = MeCab.Tagger("-Owakati")
+special_characters = '[\,\、\。\．\「\」\…\？\・\!\-\;\:\"\“\%\‘\”\�]'
+
+def norm_everything(batch):
+    batch["sentence"] = neologdn.normalize(batch["sentence"]).strip()
+    batch["sentence"] = normalizer(batch["sentence"]).strip()
+    batch["sentence"] = wakati.parse(batch["sentence"]).strip()
+    batch["sentence"] = re.sub(special_characters,'', batch["sentence"]).strip()
+    return batch
+
+ds = ds.map(norm_everything)