-
-
Notifications
You must be signed in to change notification settings - Fork 3
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.
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
Date normalization #5
Comments
Thanks for reporting this issue @c-arvind. From what I understand the issue is that: Instead of Let me see if we can do anything to normalize dates. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Hi I tried using your normalizer for help in calculating WER for my personal use case.
I have a ground truth like so:
JUNE THIRD EIGHTEEN SEVENTY ONE OBOCOCK BROTHERS BANK AT CORYDON IOWA WAS ROBBED OF FORTY THOUSAND DOLLARS BY SEVEN MEN IN BROAD DAYLIGHT
and my predicted text (using whisper base.en) like so:
June 3, 1871, Hobacock Brothers Bank at Croix d'Inneil was robbed of $40,000 by seven men in broad daylight.
The issue is that if i normalize the ground truth then it converts $40k to the numeric amount as well as '7 men' however the first date is printed as 'June 3rd' and if i normalize the predicted text then the numbers stay in numeric form
The text was updated successfully, but these errors were encountered: