Learn about great audio processing software as James Schramko interviews Georg Holzmann of Auphonic.com.
00:25 – A technical discussion with Georg Holzmann
02:23 – How Auphonic.com started
04:50 – European broadcasting standards
05:37 – Who funds Auphonic? (Why it is FREE to use)
08:51 – Auphonic settings recommended for average podcasters
11:16 – How you can podcast better
12:15 – Distributing processed files automatically to your server
13:09 – How to get started with Auphonic?
13:55 – Bonus tips for making good quality audios and videos
James Schramko here, welcome back to Superfastbusiness.com. Got a pretty interesting discussion now, this is going to be a bit technical but I got a new friend called Georg Holzmann from a company called Auphonic.com. Welcome to the call, Georg.
Georg: Hello James.
James: I’m not sure if I pronounced that correctly.
Georg: Yeah, that’s great.
James: Probably your software can fix that up somehow. I’ll tell you just a little bit of a back story. I make a lot of podcasts. I put out something almost every day and also videos and I’m not an audio technician, which would be obvious to you and most of my listeners. A month or so ago, one of my friends sent me an email and he said “Listen, don’t take this the wrong way but I have a lot of trouble listening to your interviews in the car because I use headphones and it’s not balanced. One person’s louder than the other and sometimes the microphone is a bit hot.
I thought I was sort of adjusting the levels equally as I said I’m not an engineer and he sent me a link and he said “Use this.” He said it’s funded by the Austrian government and you put in your audio and it will compute and calculate and balance and equalize and it can tell the difference between music and voice and it will turn out something beautiful at the other end. And I thought that sounds a little bit magical, I’ll try it and oh by the way it’s free.
So I loaded up my files and sure enough the end result was clean, crisp, easy listening podcasts and since that day it’s now become part of our standard operating procedure that every single audio and every single video that we make gets loaded up and then downloaded again and then put on to our site for our customers.
Georg, I’m very interested to know about Auphonic and we’ll spell that for people. It’s A-U-P-H-O-N-I-C.com, how could it possibly do this and how is it free?
Georg: Yeah, good question. We have started to work on Auphonic well actually 2 and a half years ago now, mainly because of the popular podcaster in Germany who always speaks also about the technical side behind of podcasting and he always claim that it is so difficult and you have to figure out all of this audio engineering and also on file formats and convert it correctly and all the meta data and chapter marks for example and so I thought ’Yeah actually it’s quite easy to make a little automation software for this’, and all this started with a small script just to adjust the loudness and it started very simple and then it got more complex and complex.
For example, now when you upload your file to us, we will first analyze the audio file very detailed. So for example, we’d look in the file via music parts or via different speakers or via different background scenes and with this knowledge we can just try to automate the process, all the engineer is doing manually. So for example it equals the loudness of different speakers or between music and speech and it’s also very important to treat the music in a different way than for example, speech because music has much more natural dynamic and you can’t just remove it, it would sound stupid.
We also do noise reduction. For example, we analyze where there are noise backgrounds or where are they changing and then we take small noise samples and remove the noise from the recording and also another important part is the overall loudness of the podcast so that different files or different episodes of your audio are similar in loudness. Again, all this is done automatically.
James: And it’s fast, as well. You know, what I love about it is that someone without technical knowledge like me can edit, just do basic edits. I can put an introduction, I can put an outro, I can join a couple of voice tracks together and it just spits out something of high quality. Now I understand some of this has had an application for European radio stations where they have certain requirements or standard around the noise between commercials and the broadcast, is it sort of they have to equalize things. Has that been one of the considerations?
Georg: Actually this was not. In the beginning when we started this project, these regulations did not exist yet so actually this was in the same process as we tried to develop some common loudness targets, so I was very surprised that the big broadcasting corporations actually did the same thing, or tried to build the same thing, so yeah this was a lucky coincidence.
James: Who’s paying for this? Because you know normally I’d be used to paying a subscription fee for something like this service.
Georg: Yeah, actually at the moment we have, as you already said, it’s a funded project from the Austrian stature, and the funding is still going a little bit, so we don’t have to make money now, that’s quite great because actually our overall goal is to improve the whole infrastructure for making audio on the web, so in the moment it’s still very difficult if you want to host your podcast or to make your feeds, it’s still very difficult for non-technical users and also how to record everything.
So actually we will be working on all these parts so that it will be as easy as blogging. And yeah, we of course plan to introduce, or at least now our plan is to introduce a premium model later, so that heavy users pay a little bit for it, like for instance we have here universities which are using lots of lecture recordings and of course they cannot post produce it manually because it would be so much work so you can’t do it. That’s our plan, to make a premium model so that the small podcasters can still use it for free, and the heavy users pay for it a little bit.
James: In the last few weeks, I’ve mentioned it many, many times, and everywhere I’ve posted it, in my forum, on Facebook, on Twitter, people say “Thank you so much.” I mean the majority of my audience are not technical, but even people who are technical are saying to me, “Okay, this sounds like too good to be true, I’m going to test it out myself”, and they even deliberately record out of sync files and then run it through the machine to see what happens.
Now, they come back and say “Wow, this actually works”, and to paint a picture for you how I’m recording today, I’m on my laptop, I’m travelling, I’m in Manila, in the Philippines, I’m in a hotel room that has a very noisy air-conditioner. I’m using a Logitech headset, so it’s not my normal podcasting microphone that I would use at home, and it’s going to go into your machine and come out a little bit cleaner than what it actually went in and in fact I suspect I must be putting in some of the dirtiest, worst quality tracks ever, because you even emailed me asking if you could use one of mine as an example of how good this thing is.
So I’m guessing there’s a big difference between the start file and the end file, and the one that you sent me about was an interesting one because the battery on my microphone had gone flat, and the voice recording was very low compared to what it would normally be, and then when I got the file back, it brought it back up and it equalized it and it made it sound as if everything was just fantastic and I’m still amazed with it. The fact that you’re hands on and that you actually reply and you sent me back some suggestions on the settings that I could use….
Could you talk me through what the average podcaster should be using in terms of the default settings?
Georg: The first thing maybe – it is of course, the better the input recording is, the better the output will be, so you cannot remove everything, so that’s not possible. There’s no magical machine which can make every very crappy recording totally high quality, but of course you can clean up lots of things, so in the general settings at the moment we have a few algorithms. At the moment everything is activated by default but not the noise reduction so this is one algorithm you have to activate if you want to use it, and we will also try to automatically activate it as necessary, so this will be online soon.
But at the moment, you just, if you have some static noise in the background, like when you record in a room, you always hear computer noise, and there’s always some noise in the microphone, so then you just activate this noise and hum reduction and this should remove all the noise. We have some problems, there are problems if you are in the room, there’s lots of reverb, so if the microphone is not close to your mouth then there are other problems which may not work with the noise reduction so that is why we don’t activate it by default now but if you don’t have these problems then it should be fine to activate it.
James: Right, you actually sent me an email saying, “you should tick the noise and hum and I do that now as a standard. So I think what I’m doing is pretty much using the standard set up- ticking the noise and hum and if it’s an audio, I don’t really change anything because it outputs it as an audio by default from memory and if it’s a video, I change it to export the same as the input so if I put in an mp4, it actually comes out an mp4.
So just in case the listener is wondering, yes you put a video in and you will get a video back out but the sound will be fixed and I’ve actually retro done my podcasts for some of my shows to make it easier for my listeners but the things that I like that we weren’t doing is that we can add the artwork to the audio track and as you mentioned before- chapters.
You can actually go and put chapters on a longer audio to make it easier for people to find sections of the audio and that’s something that is quite something uncommon in the podcasting community and you can also put the artist in the description and everything so that when you load up the file, it’s going to tell you what it is when you load into iTunes or something which is a great feature.
Georg: Oh well, some remarks to chapters and things so example here in the German-speaking podcasting scene, chapter marks are very common because they make really long podcasts and actually we’re behind it to get some new chapter marks be done for mp3 and ogg files, before it was just working for mp4 files and another important thing, which I think you’re not using is we can also distribute the processed files automatically to your service or to other services.
So for example if you process your file, you don’t have to download it back again, you can just put it directly in your server when you register your FTP or STP’s or SFTP’s on our servers or for example, you can automatically send it also to YouTube , or to Soundcloud or to Libsyn or wherever you want. Also archive it or order on Amazon S3 so then it’s also distributed automatically to various locations.
James: And you can just use Dropbox too right? Just pop it into Dropbox and it will work.
Georg: Of course, you could just register your Amazon S3 as an outgoing service and then the result file will be copied to S3 automatically.
James: Yes, which is fantastic. If I’m an average person making a video or an audio and I like the sound of this and I want to use it, how do we get started with Auphonic?
Georg: Yeah, you can just make an account on our webpage and just upload an audio file or we have also an iPhone app and the Android app and that is coming soon so that you can record some audio and yeah just right, it’s for free and we also very much appreciate feedback or feature requests or problems.
James: I like that, it’s such a great thing and I see you’re quite active on Twitter and I know that my community really thankful that you exist so have you got any final tips for people who are putting out audio and video or content? Things that you’ve seen other people do really well or I imagine maybe sometimes, you even listen to some of the content to get to see who is using it and how they’re using it. Any sort of cool tips you might share as being privileged access to that sort of information.
Georg: Actually, one thing which is important is how you record your audio I think. So the microphone should not be too much away. So many people just put the microphone somewhere and then five people are standing a few meters away, yeah this wouldn’t sound good and you have a lot of room in it.
James: Got it, so you put the best input in that you can and you’ll get it cleaned up and just to give you some perspective, I have someone quote me recently for $100 podcast per episode to edit it using their Adobe Editing Suite and to make it sound good and tune it all up and turn the dials but this really is something that you can use without any knowledge, whatsoever. So thank you for creating Georg and hopefully people in the jamesschramko.com community are going to get right behind it and let you know how much they appreciate it and send you a tweet or something.
Georg: Thanks a lot.
James: You’re welcome. Alright that was Georg Holzmann from Auphonic.com. It’s free, it’s amazing, check it out and also special thank you to David Pfahler for referring me to Auphonic and suggesting that I improve the quality of my podcast that was a really good suggestion and it’s going to benefit not just me but all of the listeners and anyone who’s producing audio and video content who’s now going to be using Auphonic. Thank you. Have a great day. We’ll be back with another episode shortly.
For more podcast episodes, subscribe on iTunes
Please leave your comments below