[Update] Update

custom_common_voice.py

@@ -167,21 +167,18 @@ class CustomCommonVoice(datasets.GeneratorBasedBuilder):
         data_fields.remove("audio")
         path_idx = data_fields.index("file_path")
         script_idx = data_fields.index("script")
-        duration_idx = data_fields.index("duration")
         examples = {}
 
         with open(tsv_files, encoding="utf-8") as f:
             lines = f.readlines()
             for line in lines[1:]:
-                field_values = line.strip().split("\t")
+                field_values = line.split("\t")
                 # set full path for mp3 audio file
                 audio_path = path_to_clips + "/" + field_values[path_idx]
                 script = field_values[script_idx]
-                duration = field_values[duration_idx]
                 examples[audio_path] = {
                     "file_path": audio_path,
                     "script": script,
-                    "duration": duration,
                 }
 
         inside_clips_dir = False