Translators receive source text in every condition imaginable: copied from PDFs with broken line breaks, riddled with smart quotes and non-breaking spaces, and salted with invisible characters that confuse CAT-tool segmentation. Cleaning the source before it enters the workflow saves hours of fighting the tool later.
This workflow prepares source text for translation: fix line breaks and whitespace, normalize quotes and Unicode, and remove invisible characters so segmentation and QA behave.
The workflow
1
Repair line breaks
Remove stray line breaks pasted from PDFs so sentences segment as single units.
2
Normalize quotes and whitespace
Convert smart quotes and collapse odd spacing for consistent source formatting.
3
Standardize Unicode
Normalize accented characters so terminology matches across the project.
4
Remove invisible characters
Strip zero-width characters that break CAT-tool segmentation and QA checks.
'/%3e%3ccircle%20cx='100'%20cy='100'%20r='84'%20fill='%23093e47'/%3e%3ccircle%20cx='100'%20cy='100'%20r='79'%20fill='url(%23teal)'/%3e%3cg%20transform='translate(55,55)'%20stroke='url(%23cmd)'%20stroke-width='11'%20fill='none'%20stroke-linecap='round'%20stroke-linejoin='round'%3e%3crect%20x='27'%20y='27'%20width='36'%20height='36'%20rx='5'/%3e%3ccircle%20cx='13.5'%20cy='13.5'%20r='13.5'/%3e%3ccircle%20cx='76.5'%20cy='13.5'%20r='13.5'/%3e%3ccircle%20cx='13.5'%20cy='76.5'%20r='13.5'/%3e%3ccircle%20cx='76.5'%20cy='76.5'%20r='13.5'/%3e%3c/g%3e%3c/svg%3e)