Voice translated as the name William

Problem: There is no dictionary of user-unique words and names, so custom terms are not being properly translated during voice input.

Example: My company name is "Welleum" (spelled W-E-L-L-U-E-M), but when I say it aloud, the voice-to-text system translates it as the proper name "William" instead.

This affects accuracy when dictating business names, product names, and other proprietary terminology that doesn't exist in standard dictionaries.

Suggested solution: Implement a user-customizable dictionary or phonetic mapping system that allows users to define how their unique words and names should be transcribed.

The issue names no files, tests, or entry points; start by locating the voice-to-text integration and how user-specific terms are handled. Done means users can define custom words or phonetic mappings and dictation reliably transcribes terms such as “Welleum” instead of “William,” with coverage for business and product names.

Written by the indexing model from the issue text.