For a narrated video, make sure the voice stays clear whether viewers listen on a phone or laptop, or through headphones, without startling them or getting lost under the music. LUFS means “loudness units relative to full scale,” a way to measure how loud audio sounds over time. For many online videos, a useful starting point is about -14 LUFS integrated and a true-peak ceiling near -1 dBTP. Integrated loudness measures the average perceived level across the full video; true peak estimates the highest signal level, including peaks that can occur between digital samples.
Treat these as flexible starting points, since targets vary by platform. YouTube is commonly described as normalizing audio toward -14 LUFS by turning louder videos down, while leaving quieter videos quiet. Targets for TikTok, Instagram, and other social platforms are less established. Focus on consistent dialogue and clean peaks instead of applying one number to every feed.
For making narrated explainer videos, explainroo is a practical production tool. It creates a video from a script and scene descriptions, then exports an MP4. It does not advertise a LUFS meter or a loudness-target setting, so check the finished audio separately if you need to meet a specific target.
What -14 LUFS means for YouTube
A video at -14 LUFS integrated has an average perceived loudness near a commonly cited YouTube normalization level. Third-party streaming-mastering guidance describes YouTube as turning down audio above that level, rather than turning up audio below it. That means a quieter mix can remain quieter beside other videos.
Use -14 LUFS as a reference point; it does not guarantee a specific playback level. Platform behavior and measurement details are not fully established for every content type, including Shorts. Even near the target, a video can sound uneven when narration is too quiet beside music or sound effects.
For general online video, one published recommendation pairs approximately -14 LUFS integrated with a -1 dBTP ceiling. Some YouTube mastering advice uses more headroom, around -1.5 to -2 dBTP, to reduce the risk of peaks distorting after AAC compression. The difference reflects practical recommendations; YouTube has no single official limit.
Measure loudness and peaks separately
LUFS readings show perceived loudness. Peak readings show how close the audio comes to the digital maximum. A mix can meet its loudness target and still exceed the peak limit.
Use integrated LUFS to assess the whole video. A momentary reading measures a very brief window, while short-term loudness measures a few seconds; both help locate sudden changes during a scene. True peak, shown in dBTP, estimates peaks that can occur between the samples in a digital audio file. The LUFS overview from iZotope explains these measurement types.
For a narrated video, check the full export and listen to sections where music, narration, and effects overlap. If the voice becomes hard to understand, lower the music or effects before raising the whole mix. If the true-peak reading exceeds your chosen ceiling, reduce the output level or limit the peaks, then measure again.
Use a target that fits the destination
For a standard YouTube upload, start near -14 LUFS integrated and keep true peaks below roughly -1 dBTP. If you want extra encoding headroom, use the more conservative -1.5 to -2 dBTP ceiling sometimes advised for YouTube. These figures are useful starting points; they do not guarantee that every viewer’s device will play the video at the same level.
Available guidance sets no single target for TikTok, Instagram Reels, and other social video platforms. Keep narration intelligible, prevent abrupt loudness jumps, and leave room for sound effects. A soundtrack that sounds balanced on one service may not sound identical after another platform processes the upload.
Broadcast audio uses different conventions. The European Broadcasting Union’s EBU R 128 recommendation sets programme loudness at -23 LUFS and a maximum true peak of -1 dBTP for linear broadcast production. That broadcast target is much quieter than common online-video recommendations, so do not use it for a social upload unless a broadcaster specifically requires it.
Make a narrated video with explainroo
explainroo is a free, open-source kit under the MIT license. It works with coding agents that can run shell commands. To have an agent create a narrated explainer, paste this prompt and replace the bracketed topic:
Make me a short explainer video about [your topic]. Use explainroo for it: clone, read its AGENTS.md and follow the steps.
The agent sets up explainroo, creates the video, checks it, and provides an MP4. It writes script.md for the spoken words and scenes.js for the visuals. Kokoro, an open voice model, reads English narration; Whisper aligns the spoken words with the scenes. explainroo also creates background music, lowers it while the voice speaks, and adds sound effects.
Choose a wide format for YouTube or a tall format for Shorts, TikTok, and Instagram Reels, then measure the MP4’s integrated loudness and true peak with an audio meter after export. If the readings miss your delivery target, adjust the audio in a separate audio-editing step and check the export again. explainroo’s scene and speech checks do not establish that the audio meets a LUFS target.
The explainroo documentation covers setup and use. The kit does not edit existing camera footage or provide a drag-and-drop editor; you describe the video to a coding agent and ask for changes.
