diff options
Diffstat (limited to 'egs/wsj/s5/steps/cleanup/internal/segment_ctm_edits.py')
-rwxr-xr-x | egs/wsj/s5/steps/cleanup/internal/segment_ctm_edits.py | 5 |
1 files changed, 4 insertions, 1 deletions
diff --git a/egs/wsj/s5/steps/cleanup/internal/segment_ctm_edits.py b/egs/wsj/s5/steps/cleanup/internal/segment_ctm_edits.py index b6c7e8a7c..39f6d38d6 100755 --- a/egs/wsj/s5/steps/cleanup/internal/segment_ctm_edits.py +++ b/egs/wsj/s5/steps/cleanup/internal/segment_ctm_edits.py | |||
@@ -807,10 +807,13 @@ def TimeToString(time, frame_length): | |||
807 | 807 | ||
808 | def WriteSegmentsForUtterance(text_output_handle, segments_output_handle, | 808 | def WriteSegmentsForUtterance(text_output_handle, segments_output_handle, |
809 | old_utterance_name, segments): | 809 | old_utterance_name, segments): |
810 | num_digits = len(str(len(segments))) | ||
810 | for n in range(len(segments)): | 811 | for n in range(len(segments)): |
811 | segment = segments[n] | 812 | segment = segments[n] |
812 | # split utterances will be named foo-bar-1 foo-bar-2, etc. | 813 | # split utterances will be named foo-bar-1 foo-bar-2, etc. |
813 | new_utterance_name = old_utterance_name + "-" + str(n + 1) | 814 | new_utterance_name = "{old}-{index:0{width}}".format( |
815 | old=old_utterance_name, index=n+1, | ||
816 | width=num_digits) | ||
814 | # print a line to the text output of the form like | 817 | # print a line to the text output of the form like |
815 | # <new-utterance-id> <text> | 818 | # <new-utterance-id> <text> |
816 | # like: | 819 | # like: |