Skip to content

Handle empty alignments in find_alignment (companion to the CTranslate2 divide-by-zero fix for #1342) #1460

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
MAkcanca wants to merge 1 commit into
base: master
Choose a base branch
from
Open

Handle empty alignments in find_alignment (companion to the CTranslate2 divide-by-zero fix for #1342) #1460

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
6 changes: 6 additions & 0 deletions faster_whisper/transcribe.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1717,6 +1717,12 @@ def find_alignment(
for result, text_token in zip(results, text_tokens):
text_token_probs = result.text_token_probs
alignments = result.alignments
if len(alignments) == 0:
# No alignment available, e.g. the window was shorter than the
# encoder stride so there were no frames to align against
# (ctranslate2 returns an empty alignment for such windows).
return_list.append([])
continue
text_indices = np.array([pair[0] for pair in alignments])
time_indices = np.array([pair[1] for pair in alignments])

Expand Down
Loading