Setting Music Under Narration So the Words Stay Clear

The most common complaint about narrated videos is that the voice is hard to hear, and the cause is almost never the voice. It is the music. A track that sounds balanced when you listen to it alone competes with speech the moment they play together, because both occupy the same range of frequencies and the human ear can only follow one clearly at a time.

Getting it right is a matter of a few decisions made in a sensible order: pick music that leaves room, set it low enough, lower it further while someone is speaking and check the result somewhere other than a quiet room.

Choose music that leaves room

The easiest fix is at the choice stage. Music with a prominent melody, especially one with vocals, fights with narration because the ear tries to follow both. Instrumental music with a steady, simple texture sits underneath a voice far more comfortably. Piano, soft electronic pads and gentle acoustic guitar work well. A busy arrangement with a lead instrument playing the whole time does not.

• Avoid tracks with vocals under narration.
• Prefer a steady, simple arrangement over a busy one.
• Avoid a lead instrument that plays continuously in the speech range.
• Choose a track longer than the video, so you never need to loop it audibly.

How far down to set it

A common starting point is to place the music so that it is roughly fifteen to twenty decibels below the voice while someone is speaking. That is much lower than most people expect: it sounds almost too quiet on its own and sounds right when the voice is on top. If you can follow the music's melody while the narrator is talking, it is too loud.

Ducking

Ducking means automatically lowering the music whenever the voice is present and raising it when the voice stops. It is what produces the polished feel of professional videos, where the music seems to breathe around the speech. Most editors offer it as an automatic option, and where they do not, you can draw the volume changes by hand. The key is to make the changes smooth, over a quarter of a second or so, so the movement is felt and not heard.

Equalise out the clash

If the music is still getting in the way after lowering it, a small cut in the range where speech lives, roughly one to four kilohertz, makes space for the voice without making the music sound thin. A gentle reduction of three or four decibels on the music in that range is usually enough. It is a subtle change, and it often does more for clarity than turning the music down further.

Check it where people will listen

A mix that works on studio headphones can fail on a phone speaker, where small speakers lose the low end and the music's remaining content crowds the voice. Always do a final check on a phone speaker and, if you can, in a noisy room. If the words are clear there, they will be clear almost anywhere.

Generated narration is consistent in level from start to finish, which makes mixing easier than with a recorded voice. You set the music once and it stays balanced, with no wandering voice to chase.

Licensing

Whatever the music, make sure you have the right to use it. Free to download is not the same as free to use in a video, and a platform may flag a track whatever you believe about its licence. Choose from a source that states its terms clearly, and keep a note of where each track came from.