The Converter will use fonts found in your system font dir, and in its install dir. (The Converter scans for fonts when it first starts up)
Fonts are specified in a Word document using XML like:
<w:rFonts w:ascii="Noto Sans" w:cs="Noto Sans" w:hAnsi="Noto Sans"/>
(see further, rFonts in the OpenXML spec
If the font specified is present on your system, the Converter will use it.
If the exact font is not present, but the Converter is able to find one it knows to be metrically compatible, it will use that.
If the Converter can't find a font to use, you wil see 'tofu', a square box indicating a missing font. To prevent this, you need to add an appropriate font.
First, identify the missing font. You can do this in Word, or by unzipping the docx and looking at the rFonts element (in document.xml or styles.xml)
Next, you'll need to source that font (or a substitute).
Add this font to the OS font dir or to the converter's install dir
If you have added a substitute font, edit fonts.xml (you'll find it in the converter's install dir)
For example, suppose I have a document which uses fonts SimHei and 華康中黑體 but on my computer, I only have SimSun and Arial Unicode MS available.
I would place entries like the following in fonts.xml:
<font name="SimHei"> <fallback name="SimSun"/> <fallback name="Arial Unicode MS"/> </font> <font name="華康中黑體"> <fallback name="SimSun"/> <fallback name="Arial Unicode MS"/> </font>
After editing fonts.xml, be sure to restart the pds conversion service, so it re-reads fonts.xml
Verify it is working as expected. To do this, create a short test docx using the font of interest, convert it to PDF, then open with Acrobat Reader, go to File > Properties > Fonts, and verify the font is listed.