Skip to content
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

Jump to bottom

word timing tweaks #1559

Merged
merged 3 commits into from
Aug 7, 2023
Merged

word timing tweaks #1559

Changes from 1 commit
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
95826d4
word timing tweaks
taylorchu Jul 29, 2023
9fccc7d
comment on eot
taylorchu Jul 30, 2023
b6077ba
clearer comments
taylorchu Jul 30, 2023
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
4 changes: 2 additions & 2 deletions whisper/timing.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -215,6 +215,8 @@ def find_alignment(

words, word_tokens = tokenizer.split_to_word_tokens(text_tokens + [tokenizer.eot])
word_boundaries = np.pad(np.cumsum([len(t) for t in word_tokens[:-1]]), (1, 0))
if len(word_boundaries) <= 1:
Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This fixes crashes because word_boundaries could be empty.

>>> word_tokens = [[5, 1],[3,2,1], [1]]
>>> np.pad(np.cumsum([len(t) for t in word_tokens[:-1]]), (1, 0))
array([0, 2, 5])
>>> word_tokens = []
>>> np.pad(np.cumsum([len(t) for t in word_tokens[:-1]]), (1, 0))
array([0.])

return []

jumps = np.pad(np.diff(text_indices), (1, 0), constant_values=1).astype(bool)
jump_times = time_indices[jumps] / TOKENS_PER_SECOND
Expand Down Expand Up @@ -297,8 +299,6 @@ def add_word_timestamps(
# hack: truncate long words at sentence boundaries.
# a better segmentation algorithm based on VAD should be able to replace this.
if len(word_durations) > 0:
median_duration = np.median(word_durations)
Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

f572f21#r120932778

max_duration = median_duration * 2
sentence_end_marks = ".。!！?？"
# ensure words at sentence boundaries are not longer than twice the median word duration.
for i in range(1, len(alignment)):
Expand Down