Updated loading script

speechdat.py

@@ -98,12 +98,22 @@ class SpeechDat(datasets.GeneratorBasedBuilder):
             text_line = get_single_line(text_file)
             if text_line is None or text_line == "":
                 continue
-            with open(wav_file, "rb") as wav_data:
-                yield str(wav_file), {
-                    "path": str(wav_file),
-                    "sentence": text_line,
-                    "audio": {
+            size = os.path.getsize(wav_file)
+            if size > 1024:
+                with open(wav_file, "rb") as wav_data:
+                    yield str(wav_file), {
                         "path": str(wav_file),
-                        "bytes": wav_data.read()
+                        "sentence": text_line,
+                        "audio": {
+                            "path": str(wav_file),
+                            "bytes": wav_data.read()
+                        }
                     }
-                }
+
+
+
+def normalize(text):
+    # remove ~
+
+
+    return text