Fix transcription column in AutomaticSpeechRecognition task template

tv3_parla.py

@@ -74,7 +74,7 @@ class Tv3Parla(datasets.GeneratorBasedBuilder):
             license=_LICENSE,
             citation=_CITATION,
             task_templates=[
-                AutomaticSpeechRecognition()
+                AutomaticSpeechRecognition(transcription_column="text")
             ],
         )