Fix cpWER#77
Merged
EduardoPach merged 4 commits intomainfrom Nov 21, 2025
Merged
Conversation
dbrkn
approved these changes
Nov 21, 2025
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What does this PR do?
This PR fixes the
ConcatenatedMinimumPermutationWERmetric, which would incorrectly computecpWERhypothesis_speakers < reference_speakersin the Greedy matching path, and for the Hungarian Matcher path,cpWERwould be incorrectly computed forhypothesis_speakers > reference_speakers.We spotted this while running evaluations on
callhome-english, therefore, I took the references and hypotheses and added them as tests. Since thecallhome-englishlicense doesn't allow redistributing the transcriptions, these files exist on a private HF repository.Although I was able to fix the implementation using MeetEval
cpWERimplementation as baseline (same amount of edits per word, although some edits were classified differently compared tojiwer), I decided to addmeetevalas a new dependency, as they provide a bunch of other metrics implementations. Their implementation is battle-tested, and as a bonus, they have a pretty neat visualization tool that we could consider using in the future.