Audio Mixing for Video

There is a persistent adage in filmmaking: audio is fifty percent of the viewing experience. In many formats, such as documentaries or corporate interviews, it is arguably more. A pristine 4K image paired with hollow, echoey, clipping audio feels profoundly unprofessional. A proper audio mix in video editing involves balancing dialogue, sound effects (SFX), and music to create a cohesive soundscape.

The Hierarchy of Sound

In almost all video formats, dialogue is king. If the audience cannot hear what the subjects are saying, the narrative fails. Therefore, the mix is built around the dialogue tracks. The general target for dialogue in a digital mix is around -12dB to -6dB on the audio meters. Music should sit underneath, usually ducking down to -20dB or lower when someone is speaking, and rising during montages or transitions.

Equalization (EQ) for Clarity

An Equalizer allows you to adjust the volume of specific frequencies within an audio clip. The most crucial EQ move for dialogue is the High-Pass Filter (also known as a Low-Cut Filter). Human voices generally do not produce useful fundamental frequencies below 80Hz. By applying a high-pass filter around 80Hz, you instantly remove low-end rumble caused by air conditioning, distant traffic, or microphone handling noise without affecting the voice.

Additionally, small EQ cuts in the "muddy" frequencies (around 200Hz-300Hz) can clear up a boomy recording, while a slight boost in the "presence" range (around 3kHz-5kHz) can help articulation push through a dense music track.

Dynamic Compression

Human speech is incredibly dynamic. A subject might whisper one sentence and shout the next. If you set the volume so the whisper is audible, the shout will clip (exceed 0dB and distort). If you set the volume for the shout, the whisper is lost. A Compressor solves this by automatically reducing the volume of the loudest peaks, reducing the dynamic range.

Once the loudest peaks are compressed (turned down), you can raise the overall gain of the entire track. This results in a thicker, more consistent vocal presence that sits comfortably above the background music without ever clipping.

Room Tone and Noise Reduction

Every room has a unique sonic fingerprint, known as room tone. When assembling dialogue from different takes or adding gaps to fix pacing, total silence is jarring. Editors use recorded loops of room tone to fill these gaps, masking the edits.

Modern noise reduction tools, powered by AI (such as vocal isolation plugins), are miraculous for rescuing terrible audio. However, overusing noise reduction will strip frequencies from the voice, leaving the subject sounding like they are underwater or speaking through a tin can. Noise reduction should be applied gently, prioritizing voice naturalness over absolute silence.

Loudness Standards (LUFS)

In the past, audio was measured purely in peak decibels. However, peak meters do not represent perceived loudness. Today, the industry standard is LUFS (Loudness Units relative to Full Scale), which measures average loudness over time. Delivering audio at the correct LUFS target is critical to ensure your video plays at the same volume as others on a platform.

• Broadcast Television: Typically strictly regulated at -23 LUFS or -24 LUFS.
• YouTube and Streaming: Platforms normalize audio to around -14 LUFS. If your mix is quieter than this, YouTube won't boost it, making your video sound quiet compared to the next one.
• Podcasts: Generally targeted at -16 LUFS (stereo) or -19 LUFS (mono).

A final Limiter applied to the master output bus is the last line of defense, ensuring that no stray transient peaks ever cross -1dB (True Peak), protecting the final render from digital distortion.