Resolve speaker error

chall.py

@@ -351,7 +351,7 @@ class Chall(GeneratorBasedBuilder):
                 segment_id = f"{audio_id}_{str(segment_i).rjust(3, '0')}"
 
                 segment_data["audio_id"] = segment_id
-                segment_data["speaker_id"] = segment["speaker"]
+                segment_data["speaker"] = segment["speaker"]
                 segment_data["words"] = segment["words"]
 
                 segment_data["raw_text"] = raw_text = self.get_raw_text(segment["words"])