How to Normalize Your Podcast Audio, and Why You Need To Do It

 
audio-car-controls-3102.jpg
 

You're reading a post from Backyard Media's Podcasting 101, a series of guides meant to explain podcasting and podcast advertising to new and current podcast creators. To see our other guides, click here.

New podcast creators often find that making a podcast involves a lot of steps. Some are expected, like using an audio editor to cut your podcast together. Some are less obvious, like normalizing your podcast’s audio. What does normalizing a podcast's levels, also called Loudness normalization, actually mean? Why should creators care about this process, and how exactly do they do it? Today, we’ll give a short explanation of what Loudness is, as well as a step-by-step guide for how to normalize your podcast.

What Loudness is, and what Loudness level you should aim for

Loudness refers to a set of agreed standards in radio about how loud a radio broadcast’s signal should be. There are different standards for different regions (Europe has one, so does North America, etc). The technical meaning of these standards doesn’t mean much for most podcast creators. The important thing to understand is that Loudness is measured in Loudness Units (LU), and a change in LU represents a change in an audio signal’s volume that the human ear can notice.

Compare this with the peak meter in your audio editor, which is measured in decibels (dB). Here’s a photo from REAPER, the audio editor that we use for our in-house podcast, showing the peak meter bouncing up and down as we play the episode audio:

 
db Meter.PNG
 

Different audio editors have the peak meter in different places, but your editor’s peak meter measures the decibel level of your podcast audio as it plays (usually it stays in the -18 to -6 dB range). But a decibel is a measure of electrical signal. An increase in dB represents an increase of volume as heard by your computer. In contrast, an increase in Loudness Units (LU) measures a volume change that your ears can hear. So that’s the difference between dB and LU. Now, we talked about those radio standards, but we haven’t said yet what Loudness you should set your podcast to.

Like dB, Loudness Units work on a negative scale, and we write it as “LUFS” when talking about this scale (you should think of LUFS as equivalent to LU). 0 LUFS is our maximum possible volume level. -15 LUFS is quieter than -12 LUFS, which is quieter than -10 LUFS, and so on. So here’s what Loudness you should set your podcast to:

  • If your podcast uses stereo sound (e.g. lots of background music, or ambient sound from an on-location, non-studio environment), your target should be -16 LUFS.

  • If your podcast uses mono sound (e.g. a studio-based talk show or narrative podcast, no or only occasional background music), your target should be -19 LUFS.

  • If you have a mix of mono and stereo sounds in your editor or you’re not sure which target to use, -18 LUFS is a safe middle ground that will work for most podcasts.

^Back to top^

Why should I normalize my podcast for Loudness?

So why are we normalizing our podcasts, exactly? Normalizing podcast audio makes sense if you think about it from your listeners’ perspective.

Say you produce a talk show podcast with two speakers, you and a guest, and both of you record on different microphones in different places. You edit the conversation with these two recordings on different tracks, and each track seems to be roughly the same dB level when you play it. You finalize your podcast’s audio and post it to your RSS feed. Your listeners receive it and start listening on their smartphone or in their car, and then they get really frustrated with your show!

Why? Because it turns out that your audio sounds much higher than your guest’s audio. Every time you ask your guest a question and the guest begins talking, your listeners reach for the volume dial on their smartphone or car speakers so they can hear the guest. And every time you start speaking again, they have to lower the volume back down.

This is the problem with relying only on dB for volume. In an audio editor, your podcast’s audio may seem fine, but when people listen to the final product, it sounds too loud or too quiet in parts. The Loudness standard fixes this issue because it’s designed to measure what we’re actually hearing.

Podcast creators who want to produce professional-sounding shows should make sure to normalize their audio. Sponsors want to advertise on shows that sound good, and normalizing your podcast’s volume is a crucial step to achieving a pleasant and professional sound. (We talk about this and other professionalization steps that creators should take in “A Guide to How to Get Sponsored for New Podcasts” .)

^Back to top^

The easy way to normalize your podcast audio: Auphonic

You’ve got your Loudness target in mind, now you have to normalize your podcast to it. The simplest way to normalize your audio to a specific Loudness standard is to use an automated service like Auphonic. We’ve mentioned Auphonic before in our Ultimate Tools guide. It’s a freemium audio processing service created by a consortium of German audio producers. Auphonic makes Loudness normalization as simple as a few clicks. It takes your finished-but-not-normalized podcast audio and processes it so that different voices, and even a single speaker throughout a conversation, will have a similar overall volume.

Step 1: Make a free Auphonic account

So, how do you use Auphonic? First, you need to make a free account here: https://auphonic.com/accounts/register/

Once you’ve done that and confirmed your account by email, you can start processing your audio. Auphonic gives all users 2 free hours of audio processing time every thirty days (referred to as “Recurring Credits”). If you have a 30-minute podcast every two weeks, you’re set with the free account. However, if you publish more than 2 hours of audio per month, you can buy small amounts of time at affordable rates (see the pricing table here).

Here’s what the homepage looks like:

 
Auphonic Homepage.PNG
 

Step 2: Start a New Production

To process your podcast’s audio, you need to start a “new production” by hitting the red button at the top, which takes you to this form:

 
 

Step 3: Input your processing settings

You must upload your podcast’s audio file (ideally WAV format) with the upper-right “Choose File” button.

(NOTE: Once you upload your audio file here and choose “Start Production”, Auphonic will use your audio credits. You can return to this form to adjust your settings, but you cannot change the file once you’ve started processing it. If you do change the file, Auphonic will charge you again for the audio time.)

You can add metadata tags to your audio file under “Title”, “Artist”, etc, which will show up in your computer’s File Explorer when you download the final audio (these are much like how a song in iTunes has artist and album tags). Add your podcast’s cover art with “Cover Image” to embed it into the file.

You can also add optional Chapter Markers, which allow listeners using certain podcast apps to skip between different sections of your shows’s content. Hit “Add Chapter Mark Row” and insert the starting timestamp for a section of your show’s content, and give the section a title. This is useful for listeners using apps like Overcast or Pocket Casts, and can give them a preview of what you’ll talk about in your show.

Under Output Files, set your file to MP3 and choose the bitrate, which determines the quality and size of your podcast audio file (96kbps to 128kbps is the range for most spoken word podcasts). Choose “mono” if you want to mix down a stereo sound file to mono-only.

Under Audio Algorithms, leave the Leveler, Loudness Normalization, and Filtering boxes checked. Your Loudness Target menu is where you decide what LUFS to set your podcast to - see the above section to decide between -16, -19, and -18 LUFS.

You can choose to use the Noise and Hum Reduction option if any of your audio has unwanted background noise, like an AC unit running behind an interviewee’s voice. If you do choose this tool, set the Reduction Amount to Auto and Auphonic will adjust it as needed through the file. This setting may take some tinkering, but Auphonic allows you to listen to the processed audio before downloading the file and “paying” with credits, so you will have the chance to change your processing settings. If Auto makes the sound too muffled or leaves spots without any sound, change to a manual reduction amount.

Once you’ve entered these settings, hit “Start Production”. Auphonic will take some time to upload your audio file and process it.

Step 4: Check your normalized audio and download the finished file

Here’s what Auphonic shows you when it’s finished normalizing your podcast:

Auphonic Results.PNG

Auphonic shows you a basic media player and graphical waveform where you can listen to the original audio (Input) and the processed audio (Output). You’ll be able to see a visual difference in the waveform, with the Output file being a closer level throughout. If you open up the details in the bottom right corner (as we’ve done above), you can see the measurement of your podcast in LU before and after.

Check different spots of the audio with the media player by clicking on the image and moving the red cursor, listening to make sure the audio sounds alright. You’ll notice that quieter voices are now louder and the volume is consistent throughout. If you used Noise/Hum Reduction, check to make sure the problematic background noise was removed. If you want to change your LUFS target or want to adjust the Noise/Hum Reduction Amount, you can click on “edit this production” to go back to the settings and re-process the audio. (You must do this before hitting “download", or you will be charged for the audio.)

Once you’re satisfied with the audio, choose “download” to download the MP3 file to your computer. Proceed with your production process and upload this MP3 as your final podcast file.

^Back to top^

The advanced way to normalize your podcast audio: Loudness Meters

Now, there may be a few reasons you wouldn’t want to use Auphonic. Perhaps you want a more hands-on approach to crafting the final sound of your show, and submitting your finished (but not normalized) audio to an automated service provokes some anxiety. Also, if you produce many hours of audio per month but have a smaller budget, Auphonic may become too expensive. (Again, take a look at Auphonic’s pricing for monthly Recurring Credits, which may reduce your overall cost)

You can choose to normalize your podcast with a paid plugin called a Loudness Meter. These meters integrate with your audio editor and measure your podcast’s audio in LUFS, similar to Auphonic.

There are many loudness meters out there. Two that come highly recommended in the podcasting industry are:

Note that if you use Adobe Audition CC or Hindenburg Journalist Pro, these DAWs have built-in Loudness plugins you can use.

Each of these meters look and operate slightly differently, but they use the same terminology. You’ll want to open up your Loudness Meter and start playing your audio. Pay attention to the Short-Term Loudness measurement in LUFS. This is the measurement that you need to match your Target LUFS (either -19, -16, or -18 LUFS).

See the video below for a demonstration of the WLM Plus meter using the free DAW Audacity (skip to 3:55 for the most relevant part):

 
 

Notice how WLM Plus analyzes the audio file, then reports the Short Term and Long Term Loudness (two different average LUFS ratings), as well as the Range (the difference between the highest and lowest Loudness in the podcast file). Again, pay attention to the Short Term measurement. Once you know what the difference is between your current Short Term Loudness and your Target Loudness, you can use the Gain slider to add the difference. This changes your audio, so make sure you’re finished editing your show before you use this meter.

By using a manual Loudness Meter, you can see the effect your gain adjustment has on your podcast’s normalized Loudness, and you can immediately hear how it will sound to your listener. Loudness Meters obviously mean making an upfront purchase, but they can make financial sense in the long run if you’re producing many podcast episodes.

No matter which route you take, normalizing your podcast is a must if you want your show to sound good to your listeners and seem professional to potential sponsors. Thankfully, normalizing doesn’t take too much effort, especially once you’ve gotten comfortable with the normalization process and have integrated it into your production cycle.

^Back to top^

For more on Loudness standards and the technicalities behind Loudness, check out these articles from Transom:

Want to read more guides for podcast creators? Click here to see our other guides.