Update libriheavy.py

libriheavy.py

@@ -153,7 +153,6 @@ class Libriheavy(datasets.GeneratorBasedBuilder):
                     # skip the last utterance
                     if utterance_id == sorted(list(text.keys()))[-1]:
                         continue
-                    print(utterance["phone_segments"])
                     result =  {
                         "id": chunk["speaker_id"] + "_" + utterance_id,
                         "speaker_id": chunk["speaker_id"],
@@ -163,7 +162,7 @@ class Libriheavy(datasets.GeneratorBasedBuilder):
                             {"start": segment[0], "end": segment[1], "word": segment[2]} for segment in utterance["word_segments"]
                         ],
                         "phone_segments": [
-                            {"start": segment[0], "end": segment[1], "word": segment[2]} for segment in utterance["phone_segments"]
+                            {"start": segment[0], "end": segment[1], "phone": segment[2]} for segment in utterance["phone_segments"]
                         ],
                         "mel_spectrogram": npz[str(utterance_id)].item()["mel"][0][0],
                     }
@@ -181,7 +180,7 @@ class Libriheavy(datasets.GeneratorBasedBuilder):
                         {"start": segment[0], "end": segment[1], "word": segment[2]} for segment in utterance["word_segments"]
                     ],
                     "phone_segments": [
-                        {"start": segment[0], "end": segment[1], "word": segment[2]} for segment in utterance["phone_segments"]
+                        {"start": segment[0], "end": segment[1], "phone": segment[2]} for segment in utterance["phone_segments"]
                     ],
                     "mel_spectrogram": npz[str(utterance_id)].item()["mel"][0][0],
                 }