What’s the best way to master your audio for YouTube? Should you master your audio as loud as possible? Does your source audio quality really matter if YouTube is going to re-encode it anyway? With more content being created for YouTube every day, we’re asked these questions more and more. Let’s take a look at some best practices you can employ to make sure the audio in your YouTube videos sounds as good as possible.
Start with (and Export) High-quality Audio
In the YouTube video-formatting guidelines, you are encouraged to “upload videos that are as close to the original, high quality source format as possible to increase the likelihood that your videos will play in higher quality.” The same concept applies to the audio you’re using for your video — using high-quality PCM audio, like WAV (.wav) and AIFF (.aif — Audio Interchange File Format) file formats, will fare better than compressed audio formats (like MP3 files). Starting with the highest-quality audio ensures the best results when you ultimately encode and export your final video as well as when YouTube inevitably re-encodes your video for streaming.
When it’s time to export your video, you’ll want to make sure you’ve checked the audio settings to make sure you’re exporting high-quality audio. YouTube recommends exporting your video either as the DVD-compliant MPEG-2 format or as an MPEG-4 using the H.264 video codec. For MPEG-2, YouTube recommends setting the audio codec to either MPEG Layer II or Dolby AC-3 with an audio bit rate of 128kbps. For MPEG-4, YouTube recommends selecting the AAC audio codec with an audio bit rate of 128kbps or higher.
The important thing to remember when exporting is to choose the highest audio bit rate possible — unless you need to restrict the file size for some reason, there’s no reason not to. Anything lower than 128kbps can significantly reduce your audio quality, even if you used high-quality audio files during video editing.
In case you’re curious how the Sweetwater video team exports our YouTube videos, they generally export using the H.264 video codec with the AAC audio codec and an audio bit rate of 320kbps. The audio files used in the project are usually 24-bit/48kHz WAV or AIFF files.
Aim for a Specific Loudness Target
The audio meters in your video-editing software are likely going to display dBFS (which stands for “deciBels relative to Full Scale”), which helps you assess how high audio signals are peaking but does not give you an idea of overall loudness — that’s what the LUFS (Loudness Units relative to Full Scale) metering system is for. As you’re balancing the audio during editing, using a LUFS-capable metering plug-in is crucial. You may have one included with your video editor already, and third-party options are available like the Waves WLM Plus loudness meter.
YouTube guidelines remind us to aim for a target of -13 to -15 LUFS in overall loudness (often shown as “integrated” or “long-term” loudness) and no louder than -9 LUFS for transient peaks (often shown as “peak” or “short-term” loudness). This is because YouTube normalizes audio to an extent — if your video is quieter than the recommended -13 to -15 LUFS, then YouTube will increase the volume to reach that target. And, if your video is louder than that, it will get turned down. This is a good thing — it discourages the practice of overcompressing audio to make it sound louder than the competition and encourages more dynamic audio.
LUFS Meters Will Keep You on Track
As you’re balancing the audio elements for your video, keep an eye on the LUFS value of your audio mix and bring levels up or down to get the overall loudness in the -14 LUFS range. If you see that short-term/peak LUFS values are noticeably louder than -9 LUFS, then identify the audio that is causing the peaks and apply some compression to dial the peaks back down. If your audio isn’t peaking that high, say around -11 or -12 LUFS, then it’s possible that something in your mix is being compressed too heavily.
In the end, trust your ears as to whether the mix is good or not, but use the LUFS metering to see whether the actual perceived loudness is in the preferred range or not. What you’re trying to avoid is overcompressed audio, which can be fatiguing for viewers and which will be turned down during YouTube’s re-encoding process anyway. Said another way, dynamic audio that gets turned up during YouTube’s re-encoding will generally sound much better than overcompressed audio that gets turned down.
Some of the most popular metering plug-ins (and software suites that include comprehensive metering) at Sweetwater include the NUGEN Audio MasterCheck plug-in, the Waves Content Creator Audio Toolkit plug-in bundle, and the iZotope Insight 2 essential metering suite.
Leave Some Headroom
We’ve focused on perceived loudness so far, but peaks are important, too. When YouTube re-encodes your video for streaming, that process can cause audio peaks that may be higher than the peak levels your meters are displaying. It’s recommended to leave 1dB–2dB of headroom on your master fader (you should be able to reach your LUFS target while leaving this much headroom). This way, you’ll not only ensure that the audio you export is not distorting, but you’ll also be preventing unexpected peaks created during re-encoding from distorting your audio upon playback.
A Tip for Mixing Your Audio for Video
Whether you’re creating music videos, product reviews, interviews, or any other type of content for YouTube, you want to ensure your audio will sound its best regardless of the playback system it’s heard on. Since some of your viewers will hear your audio from their mobile device, others on desktop speakers, others with headphones or earbuds, and so on, it’s best not to try to tailor your audio mix to a certain type of playback system.
One helpful technique you can use to make sure your overall audio mix is well balanced is to listen to it at a very low volume level and check that you can still hear the most important audio elements. When mixing music, this would mean making sure you can still hear a balance of the vocals, lead instruments, and rhythm section even at very low volume. The same concept applies to video — decide which elements are crucial to the audio mix (dialog, narration, music, sound effects) and make sure they have a good balance at low volume. They’ll maintain their balance at higher volumes, and this helps ensure that balance is maintained regardless of the volume listeners prefer.
Summary
- Use a LUFS metering plug-in and aim around -14 LUFS for your overall audio mix
- Leave 1dB–2dB of headroom on your master fader
- Export at a high audio bit rate, 128kbps at the minimum
- Use high-quality WAV or AIFF audio in your editing session; avoid MP3 and other compressed formats
- Listen to your audio mix at a very low volume level to ensure all the important elements are balanced
It’s Better to Get Turned Up Than Turned Down
With all this focus on LUFS values and headroom, it’s easy to miss the good news in all of this — the loudness wars are effectively over on YouTube. You don’t need to mix your audio as loud as possible — YouTube will simply turn up your audio if it’s a bit quiet (good) and turn it down if it’s too loud (bad). So, aim for the recommended loudness target and focus on having great-sounding audio files to work with, and you’re already well on your way. And, if you find yourself struggling to achieve great-sounding YouTube videos with these tips, consider giving your Sweetwater Sales Engineer a call at (800) 222-4700. Many of us are content creators ourselves — all of us are picky about audio quality — and we’ll be eager to help you refine your process.