Fixing (1) HierarchicalBehavioralTable.add_interval kwargs handling a…

…nd (2) issue with hyphens in syllables_data_extractor and timitsounds_converter (by switching to comma)

src/pynwb/ndx_hierarchical_behavioral_data/hierarchical_behavioral_data.py

@@ -49,10 +49,10 @@ def __init__(self, **kwargs):
             allow_extra=True)
     def add_interval(self, **kwargs):
         # automatically populate the time with the start time of the first element of the next tier
-        if getattr(kwargs, 'start_time', None) is None:
+        if kwargs.get('start_time', None) is None:
             kwargs.update(start_time=self['next_tier'].target.table['start_time'][kwargs['next_tier'][0]])
 
-        if getattr(kwargs, 'stop_time', None) is None:
+        if kwargs.get('stop_time', None) is None:
             kwargs.update(stop_time=self['next_tier'].target.table['stop_time'][kwargs['next_tier'][-1]])
 
         super().add_interval(**kwargs)

src/pynwb/ndx_hierarchical_behavioral_data/transcription_io.py

@@ -93,7 +93,7 @@ def syllables_data_extractor(syllables_phonemes_data):
     for i in uniq_syl:
         x = syllables_phonemes_data[syllables_phonemes_data['syllable_number'] == i].reset_index()
         data = [
-            [x['start_time'].iloc[0], x['stop_time'].iloc[-1], '-'.join(x["phonemes"]), x['syllable_number'].iloc[0]]]
+            [x['start_time'].iloc[0], x['stop_time'].iloc[-1], ','.join(x["phonemes"]), x['syllable_number'].iloc[0]]]
         uniq_row = pd.DataFrame(data, columns=['start_time', 'stop_time', 'label', 'syllable_number'])
         syllables_data = syllables_data.append(uniq_row, ignore_index=True)
     last_row = pd.DataFrame([[syllables_data['stop_time'].iloc[-1], syllables_data['stop_time'].iloc[0], 'h#', 0]],
@@ -250,7 +250,7 @@ def timitsounds_converter(phonemes_data, syllables_data, words_data, sentences_d
     syllables = HBTSyllables(lower_tier_table=phonemes)
     cum_phonemes_count = 0
     for ind in syllables_data.index:
-        phonemes_count = len(syllables_data['label'][ind].split('-'))
+        phonemes_count = len(syllables_data['label'][ind].split(','))
         tier_ind = list(range(cum_phonemes_count, cum_phonemes_count + phonemes_count))
         cum_phonemes_count = cum_phonemes_count + phonemes_count
         syllables.add_interval(label=syllables_data['label'][ind],