Formatting - Corruption
Letters or characters are not displaying correctly. Symbols or spaces may be appearing in place of intended characters. Diacritical marks or accents may be missing throughout the file.
If you are repurposing timed text assets be sure to take care if you convert from one format to another. If for example, you have received a .PAC file and need to convert to .STL, be sure to check that no issues have been introduced during the conversion.
You can attempt to do this by checking for characters that could indicate corruption such as < > % @ # ^* ☐ (among others).
Be sure to double-check that your timed text software is correctly encoding any special flags/tags correctly. For example, italics is usually signified with <i>. However, in some instances, the tag is not encoded correctly and instead appears within the subtitle event itself.
If you are dealing with languages that include diacritical characters, perform a search for them to ensure they are still present after any format conversion.
If you are authoring Japanese subtitles, please be cognizant of the number of ruby characters used in a given container when you intend to use another ruby container in an adjacent "base character sequence". Two directly adjacent containers containing multiple ruby characters can result in those ruby characters overlapping, which would result in corruption.
Lastly, ensure the correct encoding type has been used for the timed text file.
For TTML assets ensure UTF-8 encoding is used.
For LambdaCap assets ensure Shift JJIS, UTF-8 (BOM), or UTF-16 is used.