Problems of preparing IEMOCAP dataset

Dear PiotrSobczak:

Thank you for your wonderful work!

Recently I am trying to create Speech Emotion Recognition system by following your instruction, but I encounter some problems of preparing IEMOCAP dataset.

As is described in your README, I downloaded IEMOCAP dataset and create dataset pickle by running your https://github.com/PiotrSobczak/speech-emotion-recognition/blob/master/speech_emotion_recognition/iemocap_utils/mocap_data_collect.py

After that, 'data_collected.pickle' (not 'iemocap_balanced pickle') is created in the data folder. Then I changed the filename to 'iemocap_balanced pickle', and tried to create balanced iemocap by using the function in your data_loader,py

speech-emotion-recognition/speech_emotion_recognition/data_loader.py

Line 153 in 8ecbcf1

def create_acoustic_dataset(**kwargs):

But I got this error:

Traceback (most recent call last):
  File "data_loader.py", line 236, in <module>
    create_acoustic_dataset()
  File "/home/xuchutian/speech-emotion-recognition/speech_emotion_recognition/utils.py", line 9, in timed
    result = method(*args, **kw)
  File "data_loader.py", line 158, in create_acoustic_dataset
    iemocap = pickle.load(handle)
ValueError: unsupported pickle protocol: 5

May I ask how to solve this problem? Thank you very much for your help.