Consider using a Table instead of the Comments function. The left column could hold your transcription and the right column your comments, as text. The right column could be formatted differently if you want. Tables do not need to have borders so it would not need to appear any different from the way you are showing now.

The comments feature is intended to mark up text, not to present text.

You could also use a program like SnagIt to create images that hold your comments and insert those.