WRITTEN BY Kyle stockbridge
Office administrator OF BOOM BOX POST
I recently sat in on a predub session with re-recording mixer, Jacob Cook. He was mixing a super heavy sound effects portion of an episode and I noticed he would regularly glance over at his loudness meter. See, he had the mix in the ballpark, but for this pass he was punching up certain scenes, making more creative choices and giving it a more dynamic, story-driven polish. Now although the decision making of his mixing was first and foremost informed by the story he was telling, he had to make sure he was still working within the guidelines of the delivery loudness specifications. This prompted me to look a bit deeper into loudness metering to help myself understand it a bit more for when I mix my own projects. Loudness can seem technical and intimidating at first, but what I’ve gathered is that it’s really about one simple idea: making sure your mix sounds balanced and consistent wherever it’s played.
LKFS/LUFS
If you’ve ever felt confused by LUFS, LKFS, or loudness specs in general, you’re not alone!
What are LUFS/LKFS? Without getting too technical, they both measure and express the perceived loudness of your mix in relation to full scale. Both use the same algorithm from ITU-R BS.1770, so they’re essentially the same and interchangeable (–24 LKFS = –24 LUFS).
LKFS = Loudness, K-weighted, relative to Full Scale.
LUFS = Loudness Units relative to Full Scale.
Okay so what’s the difference? To keep it simple: it’s mostly regional and historical:
LKFS appears in U.S. broadcast specs (ATSC A/85).
LUFS appears in European specs (EBU R128).
K-weighting refers to the frequency weighting curve used in the measurement. Although we only see the “K” in LKFS, both LKFS AND LUFS use K-weighted measurement. It mimics the way human hearing perceives loudness. It slightly emphasizes the mid and high frequencies (where our ears are the most sensitive) and attenuates very low frequencies. This makes the loudness value reflect a more perceived volume, not just an electrical signal level.
Types of loudness Measurements
Okay so you’ve just installed a loudness meter like Youlean Loudness Meter 2 and slapped it on your master track. But what do these different readings even mean?
Short-Term Loudness: A rolling average of loudness over the past 3 seconds. This shorter window of measurement is useful to the mixer for visualizing the loudness consistency from moment to moment. Measured in LKFS/LUFS
Integrated Loudness: the overall average loudness of the ENTIRE mix (or selection) over time. This is one of the most important measurements used for meeting delivery specifications. Measured in LKFS/LUFS
True Peak Max: Measures the highest actual signal peak of the mix that you’d hear after digital to analog conversion. This is expressed in “dBTP” or “decibels True Peak.” This is the most important measurement for ensuring your mix hasn’t exceeded the maximum digital signal level of 0dB, which would create clipping and digital distortion artifacts… yucky. (Check out Ian Howard’s post to see how you can use a limiter as a safety net against this!)
For more detailed information on how the math for these measurements are calculated check out this link on the Youlean Loudness Meter website!
Applying it to your mix
So let’s say you need to meet the following delivery specs for your mix:
“–24 LKFS,(+/–2), True Peak ≤ –2.0 dBTP”
What does this mean?
After metering the full mix, the overall loudness measurement (“Integrated”) is allowed to land within a range of -22 to -26LKFS (a range of 2dB above and 2dB below the specified -24dB average).
After metering the full mix, the momentary signal peak should never exceed –2.0 dBTP (“True Peak Max”)
Audio delivery specs are always evolving because content and how we consume it is always evolving (check out Jeff’s post about panning with the iPhone viewer in mind). Remember, the goal of employing loudness specs is to help ensure consistency from program to program across a broadcast or streaming platform! Different streaming platforms such as Netflix, HBO, Youtube etc. have their own ideologies around loudness and what they think is best for their viewers. With that being said, let’s talk about one more loudness spec you may be asked to meet: dialogue-weighted loudness spec.
Integrated vs. Dialogue-Weighted Loudness
Integrated: how loud the entire mix is on average over time.
Dialogue-weighted loudness: how loud the program’s dialogue is on average over time, ignoring music and effects.
Both are measured in LKFS/LUFS, but they tell us slightly different things about how a mix will feel to the audience.
Integrated Loudness
This measurement looks at the whole mix: dialogue, music, backgrounds, explosions, silence, etc… if it’s in the mix, it contributes to the calculated average loudness value. This is useful for technical consistency: if two programs both measure –24 LKFS integrated, they’ll theoretically be perceived as having extremely similar average energy level over time. However, the human ear doesn’t perceive all sounds equally. In film and television, we anchor our sense of loudness to voices. If the dialogue feels too quiet, the audience will likely think the mix as a whole is too quiet, even if the explosions and music are perfectly balanced on a meter.
Dialogue-Weighted Loudness
Dialogue-weighted (sometimes called dialogue-gated) measurements focus only on the parts of the mix that contain spoken dialogue.
It’s based on the idea that our perception of program loudness is tied to how loud the voices are, not necessarily the overall mix. By measuring only those moments, mixers can creatively “normalize” content so that dialogue plays back consistently between shows, movies, or commercials… even if the music or effects have wildly different dynamics.
For example:
A documentary might be mostly gentle background music and spoken narration while an action film might have wide dynamics of quiet scenes and then loud explosions.
Even though the action film’s integrated loudness might measure higher, a dialogue-weighted approach ensures the voices are the same perceived loudness in both — so you don’t need to grab the remote.
Why It Matters for Viewers
The difference directly affects how natural and comfortable a mix feels:
Integrated loudness ensures the overall energy of programs is controlled: good for broadcasters.
Dialogue-weighted loudness ensures the spoken word feels balanced from one title to another: good for viewers.
When you switch between two shows and the dialogue suddenly feels louder or quieter, it’s often because one was normalized using integrated loudness and the other was normalized using dialogue-weighted loudness.
This is why some streaming platforms, like Netflix, use dialogue-weighted loudness as their primary QC metric: it prioritizes what the audience actually listens to: the story being told through the voice.
Now this is a lot of information and there are a lot of numbers to pay attention to… but do not let it inhibit your creative choices while mixing! Loudness standards exist to make the listening experience of films, shows or even music more consistent and enjoyable no matter where or how it’s played back. Use your loudness meter as a tool to meet delivery specifications, but always do your best to serve the story and make your mix sound the way YOU want.
Hopefully this post has helped demystify some of the terminology and concepts behind loudness so you can approach your mixes with a little more confidence and clarity!
Which loudness meter do you prefer? Do you have thoughts on dialogue weighted vs integrated?
If you enjoyed this blog, check out these:
Lunch and Learn: Meeting Network Loudness Specifications
Lush Sound Design: Using Frequency Perception as a Tool
Lunch and Learn: Objective Based Mixing