The -24 LKFS Standard and the Loudness Range: How Audio Mixing Controls Volume in Broadcast and Streaming

The audio mixer sits in a calibrated studio, monitoring a film soundtrack at a reference level of 79 decibels, crafting a dynamic arc that sweeps from a barely audible whisper to a concussive explosion. The viewer sits in a noisy living room, watching that same film on a tablet, frustrated that the dialogue is buried and the action sequences are deafening. That tension—between theatrical intent and consumer reality—is the central conflict of modern audio engineering. It is a conflict mediated entirely by a set of invisible algorithms and strict numerical targets that dictate exactly how loud a piece of media is allowed to be.[7]

The modern era of loudness regulation began as a legislative response to a universal annoyance: the blaring television commercial. Before 2010, broadcasters measured audio using peak meters, which tracked the absolute electrical voltage of a signal. Advertisers quickly realized that by heavily compressing their audio, they could keep the signal constantly pinned just below the maximum legal peak. The result was a commercial that registered as technically compliant on a peak meter, but sounded aggressively louder to the human ear than the television show it interrupted.[3]

To solve this, the industry had to abandon peak measurement and figure out how to measure human perception. The breakthrough arrived with the International Telecommunication Union's publication of ITU-R BS.1770. Now in its fifth revision (published in November 2023), this document defines the global algorithm for measuring perceived loudness. It introduces a concept called "K-weighting," a mathematical filter that mimics the human ear by emphasizing the upper-mid frequencies where our hearing is most sensitive, while ignoring deep bass that we feel more than we hear.[6]

The ITU-R BS.1770 algorithm takes a soundtrack, runs it through the K-weighting filter, and spits out a single number representing the average perceived volume over the entire duration of the program. This measurement is expressed in Loudness Units relative to Full Scale, abbreviated as either LUFS or LKFS. A change of one LUFS is exactly equal to a change of one decibel, making the scale intuitive for audio engineers. But while the ITU provided the ruler, it did not set the speed limit. That task fell to regional broadcast authorities.[6]

In the United States, the Advanced Television Systems Committee (ATSC) published the A/85 Recommended Practice, which established a strict target: all television programming must average -24 LKFS. This standard was given the force of law when the Federal Communications Commission implemented the CALM Act. As the FCC plainly states, "The CALM Act requires commercials to have the same average volume as the programs they accompany." Across the Atlantic, the European Broadcasting Union (EBU) published its own standard, R128, which set a nearly identical target of -23 LUFS.[1][2][3]

However, a single average number creates a new problem. A film consisting of 50 minutes of absolute silence and 10 minutes of deafening explosions might mathematically average out to a compliant -24 LKFS, but it would be unwatchable. To prevent this, the EBU introduced a supplementary metric in document Tech 3342 called the Loudness Range (LRA). LRA measures the statistical distribution of loudness throughout a program, quantifying the distance between the quietest narrative moments and the loudest peaks. A live news broadcast might have a narrow LRA of 3 LU, while a feature film might demand an LRA of 15 LU or more.[4]

As audiences migrated from traditional broadcast television to on-demand streaming, the rulebook fractured. Streaming platforms are not bound by the FCC's CALM Act or European broadcast regulations, allowing them to establish their own proprietary targets. Netflix, for example, requires audio mixers to deliver a significantly quieter target of -27 LKFS, with a strict tolerance of plus or minus 2 LU. By setting a lower average target, Netflix provides mixers with more "headroom"—the space between the average volume and the maximum digital ceiling—allowing for a wider, more theatrical dynamic range.[7]

Crucially, Netflix and other major streamers measure this target differently than broadcasters do. While the ATSC A/85 standard measures the integrated loudness of the entire program, Netflix requires a "dialog-gated" measurement. As outlined in the company's Sound Mix Specifications, the -27 LKFS target must be met by measuring only the portions of the audio track where human speech is present. "Average loudness must be -27 LKFS +/- 2 LU dialog-gated," the specification reads. "Peaks must not exceed -2db True Peak."[7]

This dialog-gated approach ensures that the most critical element of a narrative mix—the spoken word—remains consistent from show to show, regardless of how many explosions or silent pauses surround it. If an engineer mixes a quiet indie drama and a loud action blockbuster, the dialogue in both films will play back at exactly the same volume in the viewer's living room. The action film will simply use the available headroom to make its sound effects louder relative to that anchored dialogue.[7]

The divergence between broadcast and streaming targets forces modern audio post-production facilities to generate multiple distinct masters for the same piece of content. A film might require a dynamic -27 LKFS dialog-gated mix for its streaming premiere, a compressed -24 LKFS program-integrated mix for its US cable broadcast, and a -23 LUFS mix for European distribution. Engineers must carefully manage their compressors and limiters—often keeping attack times between 2ms and 15ms—to hit these precise mathematical targets without introducing audible pumping artifacts.[1][2][7]

The standards themselves are continually evolving to keep pace with new consumption habits. On July 15, 2026, the ATSC announced a major revision to the A/85 Recommended Practice. The update modernizes the industry's loudness guidance to account for the rise of immersive, object-based audio formats like Dolby Atmos, and addresses the complexities of delivering consistent audio to mobile devices and smart speakers. As the ATSC noted, the revision ensures that the foundational principles of loudness management remain relevant in a fragmented hardware landscape.[5]

The mathematics of LKFS and LRA operate entirely behind the scenes, invisible to the audience. Yet they represent one of the most successful technical consensus efforts in modern media history. By shifting the industry's focus from the electrical limits of a wire to the perceptual limits of the human ear, these standards ended the era of the blaring commercial and established a baseline of acoustic consistency. The challenge for the next decade will be maintaining that consistency as the distance between the calibrated mixing stage and the consumer's earbuds continues to widen.[2][6]