Extract_features problem

[sl403]

Hi,

When I try to run experiments/InterSpeech_2015.py.
In the stage2 "Extracting features...", I get the following error.
Anyone knows how to solve this problem? Thank you for your time.

Error message: 
loading in timings file /usr/deep_disfluency/deep_disfluency/experiments/../data/disfluency_detection/switchboard/swbd_disf_train_data_timings.csv ...
992 speakers with timings input
Traceback (most recent call last):
  File "/user/deep_disfluency/deep_disfluency/feature_extraction/save_feature_matrices.py", line 267, in <module>
    lmfeatures_dir=args.lm_feature_folder
  File "/usr/deep_disfluency/deep_disfluency/feature_extraction/save_feature_matrices.py", line 138, in save_feature_matrices
    word_ix = word_2_idx_dict.get(lex_data[i])
TypeError: unhashable type: 'list'
{'asr_WER': None, 'POS_accuracy': None}