<aside> 🛠

Normalize a Video's Loudness to Broadcast Standards — 2-Pass Loudnorm + True-Peak Limiter

The final audio mastering step before publish. Audio Finish runs a two-pass loudness normalization (EBU R128 style) on the input video's audio track and applies a true-peak limiter to prevent lossy-codec clipping. Defaults hit the -14 LUFS / -1.5 dBTP target that YouTube, Instagram, TikTok, and other social platforms normalize to — meaning your uploads sound consistent and at correct volume without the platforms silently attenuating them. Zero main-view controls; targeted knobs live in Advanced Options.

</aside>

What it does

Audio Finish takes a single video, analyzes its audio track in a first pass to measure integrated loudness and true peak, then in a second pass applies gain corrections and a peak limiter to hit the target loudness (default -14 LUFS) while keeping true peaks below the ceiling (default -1.5 dBTP). The result: audio that meets modern streaming loudness norms, sounds even against other content on the same platform, and doesn't distort in lossy codecs like AAC / MP3.

The technical shape: 2-pass loudnorm (measure → correct) followed by a true-peak limiter to catch any inter-sample overshoot before it becomes audible clipping. This is the same approach mastering engineers and broadcasters use — automated into a single node.

Problem it solves

• Meet platform loudness targets – YouTube, Spotify, TikTok, and Instagram normalize incoming audio to ~-14 LUFS; pre-normalizing to that target means your upload plays at intended volume instead of being attenuated
• Consistent volume across a series – When every clip in a batch is normalized to the same LUFS target, viewers don't have to adjust volume between clips
• Prevent lossy-codec clipping – The true-peak ceiling accounts for reconstruction peaks that appear when digital audio is encoded to MP3 / AAC — headroom of -1.5 dBTP keeps intersample peaks safely below 0
• Professional mastering feel – Two-pass loudnorm produces cleaner, more musical results than single-pass automatic gain control
• Broadcast-standard alternative – Adjust the target LUFS to -23 (EBU R128) or -24 (ATSC A/85) for TV / radio delivery
• Post-narration polish – After Video Narrate, Narration Mux, or Video Concat introduces new audio, Audio Finish evens everything out into one cohesive audio experience
• Zero-tuning defaults – The main view is empty because the defaults (-14 LUFS / -1.5 dBTP) work for 90% of social publishing use cases

Input/Output

<aside>

• Input: Single video
  • Format: Standard video with an audio track
  • Source: Typically the output of Video Concat (joined final video), Narration Mux (Cold Open) (narrated single clip), or any other single-video producer </aside>

<aside>

• Output: Single video with normalized audio
  • Format: Same as input (video passes through unchanged)
  • Audio: Gain-corrected to the target LUFS and peak-limited to the true-peak ceiling
  • Length: Unchanged
  • Sync: Unchanged — audio processing is level-only, no time shifts </aside>

Configuration Options

• Target integrated loudness in LUFS. -14 is the social/YouTube norm; lower is quieter.
  • Default: -14 LUFS
  • UI: Slider (in Advanced Options modal)
  • Purpose: The target integrated loudness the two-pass loudnorm aims to hit. Lower (more negative) values produce quieter output; higher values produce louder output.
  • Reference points:
    • 14 LUFS (default) – YouTube / Spotify / Apple Music / Amazon Music — the modern streaming standard. Also fits TikTok, Instagram Reels, Facebook Reels.
    • 16 LUFS – Podcast platforms (Apple Podcasts, Google Podcasts)
    • 18 LUFS – Legacy podcast / audiobook targets
    • 23 LUFS – EBU R128 (European broadcast standard)
    • 24 LUFS – ATSC A/85 (US TV broadcast standard)
  • Note: If you master too loud (say, -10 LUFS) for a streaming platform, that platform will just turn it down — you gain nothing and often lose dynamic range. Match the platform's target for the best result.

• True-peak ceiling in dBTP. -1.5 leaves headroom so lossy-codec inter-sample overshoot stays below 0.
  • Default: -1.5 dBTP
  • UI: Slider (in Advanced Options modal)
  • Purpose: Sets the true-peak ceiling for the limiter. Prevents any audio sample (including reconstruction peaks between samples) from exceeding this level.
  • Reference points:
    • 1.0 dBTP – Aggressive; leaves minimum headroom but maximum perceived loudness
    • 1.5 dBTP (default) – Recommended for lossy-codec delivery (MP3, AAC, streaming). Standard headroom that prevents inter-sample overshoot in the encoded output.
    • 2.0 dBTP – Conservative; safer for chains that will be re-encoded multiple times
    • 3.0 dBTP or lower – Very conservative; use only if downstream processing is unpredictable
  • Why this matters: Digital audio at 0 dBFS can produce inter-sample peaks above 0 when converted back to analog or re-encoded to a lossy codec. Those overshoots are the classic source of audible distortion on encoded uploads. -1.5 dBTP headroom prevents it cleanly.