Podcasting 101: How Are My Levels?
By Alex C. Telander
One of the most inviting aspects to making your own podcast is that there’s not really a rule book to follow. There’s no publishing house vetting what type of podcast you want to make, whether you’re doing it right or wrong, and the audio drama community is welcoming and incredibly inclusive and supportive.
So you’ve got your well written script, a variety of impressive sound effects, some great original music, and your talented cast of voice actors spread around the world all with their own microphones. The first episode you’ve just finished putting together sounds great and you feel you’re ready to hit the upload button and make your podcast a reality . . . except you just did a final listen and you noticed not everything is at the same volume.
You’re using three actors in this episode and in the dialog scenes one person sounds louder than the other; the music is too quiet, and the sound effects are too loud. You just want everything to be at the same level. You’re not really sure how to do this, and definitely not without it taking a long time . . . and you’re exhausted. You’ve been working your butt off the whole last week and you’ve got to be in to work early tomorrow. So what can you do? Enter Auphonic.
Auphonic Productions is an online company that has made normalizing and equalizing the many parts of your audio file as easy as a few clicks or taps, a short wait, and then downloading your perfect sounding audio file. And the real awesome thing about it is it’s basically free.
Auphonic was started by founder Georg Holzmann when he was at university, working on audio processing, signal processing, and audio engineering. He had been listening to a lot of podcast creators telling him the most difficult part is always the audio post-production. Holzmann started working with a podcast creator, looking to ‘automate his work flow’ and make the post-production easier. He worked on creating elaborate algorithms, especially with [the] adaptive leveler, which matches the loudness of different parts of a podcast, as well as developing other tools involving encoding, tagging, and the distribution of the podcast.
Holzmann decided to make a product out of it and applied for a grant in Austria, where he’s based, and formed a team. The first version of Auphonic was released in 2012 and was originally completely free, but once the grant expired, it was necessary to introduce a pricing model.
The beauty of Auphonic is you can create an account for free, and by doing so you automatically get two hours of free audio processing time per month. There’s a lot you can do with the many settings of this free account, but I’m going to concentrate on a quick and easy overview for now.
The goal of Auphonic is to automate all the audio post-production once you have your music and dialog recorded, as well as making the steps after the audio recording is complete easy and efficient. The adaptive leveling algorithm classifies different segments for different speakers, as Holzmann explains, creating a balance between the speech and music parts, and when the two are playing together, “otherwise one would destroy the inner dynamic of the music.”
Auphonic is looking to make it so none of the post-production audio work needs to be done manually. It has also expanded its capabilities to include working with audio not just in podcasting, but also with radio stations, conference recordings, lecture recordings, and even television and film. Pretty much wherever dialog is mixed with music, Auphonic can be used.
Ever since its release, Auphonic has become a popular audio editing program to use, whether it’s with the online version, or the desktop app version. Users range into the hundreds of thousands.
Auphonic Quick and Easy
So you’ve got your free account all set up, you’re ready to make the most of your two hours of processing time, and you just want to make that first episode sound balanced and even. Where do you start?
You go into the New Production menu. Choose what file you wish to upload. Make sure ‘Leveler’ and ‘Loudnorm’ boxes are checked, choose your Loudness Target – there’s a setting for Podcasts and Mobile – and hit Start Production. And that’s pretty much it. You’ve done the hard work. The next screen shows an upload bar, as it uploads your audio file, and depending on its length this can take a couple minutes at most.
Then you wait a little longer. You can close the screen if you want and you’ll get an email when you’re audio file is all ready to download. A few minutes later you’re following the link to the page with a clear download link for your audio file. And below that is a cool looking graph showing where the volume was raised and lowered in your audio file. You can even click on an option to see the original version superimposed over the new one to show what was changed.
That’s it. Download your file and you’re ready to go. You can relax, knowing you’re audio now sounds leveled and great.
Auphonic’s Advanced Parameters
There’s a lot more you can do with Auphonic depending on your experience level and comfort with audio processing. The beauty of the program lies in the interface: you have the option of a very simple and clean layout that’s quick and easy to use, or a much more in depth display where you can customize exactly how you want your audio processed.
Firstly, your file can be either audio or video (here is a list of all the supported audio and video file types for Auphonic).
When uploading your file, you can choose to upload your audio via your computer or device, or via a website. You can choose a specific intro/ outro to add to the beginning/ end of your audio file through the same upload methods.
The next category is Basic Metadata, where you can choose the title for your audio file and select a file image to upload that will be attached to it permanently, along with details such as Artist, Album, and Track. Extended Metadata lets you create a subtitle, choose a genre and year, and create a Summary. After that you can add details about the publisher, URL, license, license URL, and Tags. You even have the option to create your own license.
Next section is Chapter Marks, where you can either import specific chapter marks that you already have in your audio file, or add your own. Once this has been done you can choose what format you want the final version of your audio file to be, along with choosing the Bitrate, adding a Filename Suffix if needed, as well as a few more tweaks.
There is also a handy section on Speech Recognition, which offers affordable speech recognition in 80 languages. You can register a speech recognition service and get it added easily; more can be found out about that here. Publishing/External Services such as Dropbox, YouTube, SoundCloud, and Libsyn (to name a few) can also be registered and added.
The final section, and perhaps most important, is the Audio Algorithms section. There are six options to play with in this category; the first four can be checked or unchecked, while the last two can be adjusted with a drop-down menu.
Adaptive Level: Corrects level differences within one file between speakers, music and speech, etc. to achieve a balanced overall loudness.
Filtering: Classifies the lowest wanted signal (male/female speech, base in music, etc.) and adaptively filters unnecessary/disturbing low frequencies in each audio segment.
Loudness Normalization: Adjusts the global, overall loudness to the specified Loudness Target (using a True Peak Limiter), so that all processed files have a similar average loudness.
Noise and Hum Reduction: Classifies regions with different backgrounds and automatically removes noise and hum in each region.
Loudness Target: Set a loudness target in LUFS for Loudness Normalization, with higher values resulting in louder audio outputs.
Reduction Amount: Maximum noise and hum reduction amount (in dB), with higher values removing more noise. In Auto mode, a classifier decides if and how much noise reduction is necessary (to avoid artifacts).
When I’m using Auphonic for adjusting audio files for Ostium I have all four options checked, with Loudness Target set to -16 LUFS (Podcasts and Mobile) and the Reduction Amount set to ‘Auto.’ And when I download the edited audio file, it sounds just perfect.
If you find that the two free hours per month for your audio work is not enough, there are two main options you can choose to add more time: you can pay a monthly, recurring fee of $11 for 9 hours, or $23 for 21 hours, or even higher totals of hours depending on what works best for you. Alternatively, you can get one-time credits of 5 hours for $12, 10 hours for $22, and a number of higher options. And if you’re looking for some Auphonic software to acquire and download, such as desktop apps, there are options for that too, which can be found here.
For the future, the team behind Auphonic is looking to build new leveling algorithms, with more detailed parameter settings, as well as building new desktop applications. They’re looking to develop the levelers to be more universal and not limited to just podcasts or speech audio. They are really looking to push the audio envelope: to work on improving and creating new things with audio.
Whether you’re a professional sound engineer or a complete novice when it comes to things that make noises, I definitely think you can’t go wrong with choosing Auphonic to magically make your work sound great.
PodernTimes and Alex C. Telander are not sponsored by Auphonics in any way.