Update ESLO.py

ESLO.py

@@ -160,7 +160,7 @@ class ESLO(datasets.GeneratorBasedBuilder):
                     "start_timestamp": start,
                     "end_timestamp": end,
                     "loss": loss,
-                    "contains_speech": contains_speech,
+                    "contains_speech": contains_speech == True,
                     "audio": {"path": audio_file,
                               "array": self._cut_audio(audio, start, end),
                               "sampling_rate": 16000}}