Keeping normalizer up-to-date with Whisper-normalizer for ASR#27
Merged
akshaykalkunte merged 1 commit intomainfrom Jan 13, 2026
Merged
Keeping normalizer up-to-date with Whisper-normalizer for ASR#27akshaykalkunte merged 1 commit intomainfrom
akshaykalkunte merged 1 commit intomainfrom
Conversation
akshaykalkunte
approved these changes
Jan 13, 2026
nhhoang96
added a commit
that referenced
this pull request
Apr 18, 2026
* add gpqa diamond * Update constants.py (#18) * updating turn handling for multi-turn evals * feat: Add Gemini support (#15) * add spokenwoz speech and text (#24) * add vllm configs and readme (#21) * added phonetics, speech_disorder, and speech_enhancement tasks - stil… (#22) * added phonetics, speech_disorder, and speech_enhancement tasks - still in need of full model scoring. Fixed small inconsistency bug in config by changing judge_properties to judge_settings. * Update the correct HF path for noise_detection task * updated scores --------- Co-authored-by: hoang <huuhoang.nguyen@servicenow.com> * voxtral and phi4 guidance (#25) * Keeping normalizer up-to-date with Whisper-normalizer for ASR (#27) * add gpqa diamond --------- Co-authored-by: oluwanifemibamgbose <oluwanifemi.bamgbose@servicenow.com> Co-authored-by: khyatimahajan <khyati.mahajan@servicenow.com> Co-authored-by: Khyati Mahajan <mahajan.khyati@gmail.com> Co-authored-by: Akshay Kalkunte <akshay.kalkunte@servicenow.com> Co-authored-by: Jash Mehta <jash.mehta@servicenow.com> Co-authored-by: Sidharth Surapaneni <40740959+pcsid@users.noreply.github.com> Co-authored-by: hoang <huuhoang.nguyen@servicenow.com> Co-authored-by: hoang <hnguy7@uic.edu>
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.
📌 Description
Fix the issue of post-processing before WER computation. There exists a minor deviation of post-processing (within
EnglishNormalizer) from the standardized whisper-normalizer. This PR ensures the compatibility with current standard normalizer.🔗 Related Issue(s)
🛠️ Type of Change
✅ How Has This Been Tested?
Specific unit testing on cases where bugs were reported.
Integration testing with re-run Librispeech-test-clean
Unit tests
Integration tests
Manual testing
Test Results / Screenshots (if applicable):
📸 Screenshots / Demos
📋 Checklist
🙌 Additional Notes