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

Fix Whisper shortform EOS #33917

Closed
wants to merge 4 commits into from
Closed

Fix Whisper shortform EOS #33917

Changes from 1 commit
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
6cd2c7b
Fix Whisper shortform
ylacombe Oct 3, 2024
a03ace5
[run-slow] whisper
ylacombe Oct 3, 2024
fde5514
[run-slow] whisper
ylacombe Oct 3, 2024
47b96fa
Merge branch 'huggingface:main' into fix-whisper-shortform-logprob
ylacombe Oct 7, 2024
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
10 changes: 7 additions & 3 deletions src/transformers/models/whisper/generation_whisper.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -891,9 +891,13 @@ def generate_with_fallback(
# remove all padding tokens
if seek_sequence[-1] == generation_config.pad_token_id:
num_paddings = (seek_sequence == generation_config.pad_token_id).sum()
seek_sequence = seek_sequence[:-num_paddings]
if return_token_timestamps and not is_shortform:
seek_outputs[i]["token_timestamps"] = seek_outputs[i]["token_timestamps"][:-num_paddings]
if is_shortform and generation_config.pad_token_id == generation_config.eos_token_id:
num_paddings -= 1

if num_paddings != 0:
seek_sequence = seek_sequence[:-num_paddings]
if return_token_timestamps and not is_shortform:
seek_outputs[i]["token_timestamps"] = seek_outputs[i]["token_timestamps"][:-num_paddings]
Comment on lines +894 to +900
Copy link
Contributor Author

Choose a reason for hiding this comment

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

I'm actually not sure if this fix covers all use-cases, @eustlb, would you mind do a sanity check ?

Copy link
Contributor Author

Choose a reason for hiding this comment

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

Let's not merge before being sure


# check which sequences in batch need fallback & which should be skipped
needs_fallback[i], should_skip[i] = self._need_fallback(
Expand Down
Loading