[Update] Update

custom_common_voice.py

@@ -172,7 +172,7 @@ class CustomCommonVoice(datasets.GeneratorBasedBuilder):
         df = df.dropna()
         chars_to_ignore_regex = r'[,?.!\-;:"“%\'�]'
 
-        for file_path, script in zip(df["file_path"], df["script"]):
+        for file_path, script, duration in zip(df["file_path"], df["script"], df["duration"]):
             # set full path for mp3 audio file
             audio_path = path_to_clips + "/" + file_path
 
@@ -186,6 +186,7 @@ class CustomCommonVoice(datasets.GeneratorBasedBuilder):
             examples[audio_path] = {
                 "file_path": audio_path,
                 "script": script,
+                "duration": duration
             }
 
         # inside_clips_dir = False