Difference in pdf conversion with and without container

Hi I use a containerized office to convert pdf. And I noticed that the conversion result is different than what is installed natively on windows 10. For this it uses the interface in C #. https://docs.microsoft.com/en-us/dotnet/api/microsoft.office.interop.word.document?view=word-pia

The bugs I found

• Although the fonts are placed in the C: \ Windows \ Fonts directory, a different font is used for the conversion
• Changed the distance between letters
• Changed the size of letters
  
  Does anyone know what this is caused by?