Skip to content

Add test for various length mask functions (code from Samsung, AI Center, Cambridge) #2894

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

Merged
pplantinga merged 8 commits into from
May 20, 2025
Merged

Add test for various length mask functions (code from Samsung, AI Center, Cambridge) #2894

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
8 commits
Select commit Hold shift + click to select a range
d14ed49
Add test for various length mask functions
May 2, 2025
0e718fb
Fix rounding error in make_padding_mask
May 2, 2025
6eb6828
Fix rounding error in compute_length_mask
May 2, 2025
279deb6
Better safety margin for length mask
May 19, 2025
c395669
Update test_length_mask.py
Adel-Moumen May 20, 2025
291b03e
docstrings
Adel-Moumen May 20, 2025
2e1365e
Remove max_value
May 20, 2025
a01b88f
Correct docstring
May 20, 2025
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
2 changes: 1 addition & 1 deletion speechbrain/nnet/losses.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -917,7 +917,7 @@ def compute_length_mask(data, length=None, len_dim=1):
mask = torch.ones_like(data)
if length is not None:
length_mask = length_to_mask(
length * data.shape[len_dim],
(length * data.shape[len_dim] - 1e-6),
max_len=data.shape[len_dim],
)

Expand Down
4 changes: 2 additions & 2 deletions speechbrain/processing/features.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1545,7 +1545,7 @@ def _load(self, path, end_of_epoch=False):
self._load_statistics_dict(stats)


def make_padding_mask(x, lengths=None, length_dim=1, eps=1e-8):
def make_padding_mask(x, lengths=None, length_dim=1, eps=1e-6):
"""Create a mask from relative lengths along a given dimension.

Arguments
Expand Down Expand Up @@ -1597,7 +1597,7 @@ def make_padding_mask(x, lengths=None, length_dim=1, eps=1e-8):

# Convert relative lengths to absolute lengths, then compute boolean mask
max_len = x.size(length_dim)
abs_lengths = (lengths * max_len + eps).unsqueeze(1)
abs_lengths = (lengths * max_len - eps).unsqueeze(1)

@pplantinga pplantinga May 19, 2025

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

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

Good catch here, I assume this is the cause of the failing tests?

mask = torch.arange(max_len, device=x.device).unsqueeze(0) < abs_lengths

# Add dimensions other than (batch, length) back into the mask
Expand Down
Loading