Silence intervals (i.e. noise-only areas) are stored as a dictionary in segmentation.py:
silence_intervals = {}
which is not the smartest idea since in case two areas of the same length are present, then one will be lost. Since length is at least 4 digits long, then chance for this bug emerging is very small. On the other hand, if one processes hundreds of thousands of files, then what can happen will eventually happen.
Silence intervals (i.e. noise-only areas) are stored as a dictionary in segmentation.py:
which is not the smartest idea since in case two areas of the same length are present, then one will be lost. Since length is at least 4 digits long, then chance for this bug emerging is very small. On the other hand, if one processes hundreds of thousands of files, then what can happen will eventually happen.