Update download_datasets_in_wav_or_mp3_and_create_csv.ipynb

download_datasets_in_wav_or_mp3_and_create_csv.ipynb

@@ -22,9 +22,9 @@
     "special_characters = '[♬「」？！“%‘”~♪…?!゛＃＄％＆（）＊＋：；〈＝〉＠＾＿｛｜｝～\"█♩♫』『.;:<>_()*&^$#@`, ]'\n",
     "\n",
     "dsa = (dataset\n",
-    "       .filter(lambda sample: bool(sample[\"sentence\"]))\n",
-    "       .filter(lambda sample: not re.search(char, sample[\"sentence\"]))\n",
-    "       .filter(lambda sample: sample[\"down_votes\"] == 0)\n",
+    "       .filter(lambda sample: bool(sample[\"sentence\"])) # Returns only samples with transcriptions\n",
+    "       .filter(lambda sample: not re.search(char, sample[\"sentence\"])) # Returns only samples without latin/roman alpha/numerics\n",
+    "       .filter(lambda sample: sample[\"down_votes\"] == 0) # Returns only samples with 0 downvotes (commonvoice)\n",
     "       )\n",
     "\n",
     "for i, sample in tqdm(enumerate(dsa)): # Process each sample in the filtered dataset\n",