Let me get this straight, I love Instagram. If we are looking at social media apps that are mostly used on your phone, Instagram is second to none. Design, user interface, speed, entertainment value; you name it, they nail it. Even though there are still a couple of things that I wish were different (video duration limits can be tough for a music producer) most of Instagram’s features are spot on.
That being said, there’s one thing in particular that made me scratch my head over and over again and made me wonder how a service that has over 1 billion users, many of whom are musicians, can mess up something like this. It’s the way Instagram compresses the audio of your video files.
So what’s it all about? When listening to your video before publishing it on Instagram, everything sounds fine. It sounds the way you mixed and mastered it. It sounds like it’s supposed to sound. Then you publish it on Instagram, and all of sudden it sounds like a 96 Kbps mp3 file. It sounds distorted. It sounds like crap. This goes for video posts, as well as, the beloved Instagram stories.
So, why is it so? Firstly, I thought, it was a one-time thing, because I didn’t experience the same audio issues every time I published a video. But, I quickly realized it wasn’t a hiccup in Instagram’s compression algorithm – it was Instagram’s compression algorithm. So, I digged deeper into it and compared the audio files of my videos before and after I published them on Instagram. How did I do this? I just published a video on Instagram as a post and as a story, then downloaded the video on a desktop pc and extracted the audio file. Then, I compared it to the original sound file of the video. One thing I noticed right away is that there is a strong high-cut at around 14 kHz. Check the frequency comparison:
You can also see it here in this 3D frequency spectrogram:
Besides that, the stereo-to-mono summing is probably the most obvious alteration that occurs when publishing on Instagram. See for yourself:
This mono summing process is also responsible for the distorted sound that you may experience when listening to the sound of your videos on Instagram. Why? – Because Instagram apparently doesn’t include an attenuator to attenuate the signal. So, if the sound of your video is already compressed, Instagram's mono summing can lead to an unwanted clipping in your audio material. Look at this image below. You can clearly see how there is some serious clipping going on with the audio that was published on Instagram:
It’s funny that Instagram itself doesn’t even give you any official requirements for sound in videos for posts and stories. At least, I couldn’t find it anywhere. The only thing I found was the official technical requirements for video ads which is not necessarily what we are talking about.
Even if it were the same requirements, it says that stereo is recommended. And guess what? Many of the ads I see are mono as well. So, why do they recommend stereo if the actual video advertisement will be published in mono? – I guess only Instagram knows.
So, how should we, as music producers, deal with this? How can we adjust to Instagram’s weird audio compression? Well, one thing would be to stay aware of this when mastering your track. You could boost your highs above 14 kHz and export it as a mono file. To be honest though for something like an Instagram story this sounds like way too much effort. If you don’t care about the 14 kHz high cut that Instagram does, you could also just export a separate Instagram stereo file with a master peaking at around -5 dBFS. This way you can avoid clipping of the respective audio on Instagram.
I also heard people saying that when publishing a video through a desktop app, for example, Uplet or Gramblr, the audio on Instagram will at least appear in stereo. On the other hand, I also read that tools like these compress your audio in other ways like reducing the bitrate. I haven’t tried this yet, but for those of you who like to schedule posts or post a lot through desktop applications anyways, this might be worth a test.
What I also heard people saying is that this is not because of Instagram, but rather because of the operating system that you are using on your smartphone. According to them, the audio compression, or at least the mono summing only occurs on iOS devices, and not on Android. Again, I don’t see a reason for this being the case, and if it was the case, it won’t help iPhone users at all. Not to mention that Apple should do something about this then, because at the end of the day we live in the year 2018 and guess what, smartphones are capable of replaying sound in stereo!
So, even though I don’t have a scientific answer of how to avoid this problem, I hope that this article gives you a little bit more insight into how Instagram deals with your audio material. Ultimately, most of the users don’t care that much about audio anyways and they won’t care about details like a 14 kHz high cut. However, for us audiophiles, the last thing we want to see is an automatic audio compression algorithm messing up the songs that we worked on for a long time.
“What can I do to make my vocals sound good?!” - You’ve probably asked yourself this question many times. If you are a singer or rapper, let me give you some quintessential advice that might help you.
First things first
Every voice is different. Settings that help the voice of the top-selling artist to be at the top of the charts, might do nothing to help your voice. In fact, such settings might even harm it. Keep that in mind as you read pieces of advice regarding frequency numbers, etc.
Your voice is unique. And what is unique, has to be treated as such. That’s why opinions about microphones vary so much. I will say this though – the better the vocal recordings, the easier it is to mix them properly.
Let’s say you take a picture of a sunset over Paris with an old, two-megapixel camera. It’s going to be a great picture, nonetheless. But if you try to make a poster out of it, you’ll end up with a blurry, pixelated mess. What the pixels and camera quality are to your eyes, bits and studio equipment are to your ears.
Expensive, high-end studio equipment can indeed give you a sense of what makes it expensive, or to put it correctly, what makes it different. Using it is a good way to train your ears. But never suppose that quality lies in the price, because like I mentioned above, every voice is unique and just because something is expensive doesn’t mean it makes your voice sound better. With that said, if you ever have the chance to record with different studio equipment, different mics, different workstations etc., please do so! It will give you the opportunity to consider the best arrangement for your voice.
Keep recording sessions dry. You can add every reverb, and every room ambiance you can think of with just a few clicks, but it is almost impossible to remove recorded room ambiance from your signal. So, do everything possible to keep your room dry. If you have a booth, you are probably in a good situation. If you don’t have one, try to build one (it’s easier than you think—just google “vocal booth self-made” to get some inspiration). If you don’t have the time or the money for it (you don’t need a lot), at least try to separate your recording area from the rest of your room in some way. Beware though, dry does not mean dead! Don't overuse the all so trendy absorbers in your vocal booth as they will swallow the life out of your voice. They will kill the high mids and highs which will make your voice sound dull.
Panning and Track Numbers
Everybody has a different approach to panning and the number of vocal tracks that are necessary. I’ll just tell you my opinion.
The lead vocals for verses are usually placed in the center. If you want to give your listener a certain intimacy, it’s always better to use only one vocal track. It just keeps your mix clear and it makes the listening experience better. I’m not a fan of doubling the entire verse. With all the subtle differences between the two takes – including the consonants that never get matched up perfectly – it just makes your vocals sound messy. If you want a clear lead vocal, only use one track.
The next thing I would do is record two tracks in which you double certain parts of the verse. Pan them both in opposite directions (15 to 40), and reduce their volume. You have to hear a difference between the doubled part and the part without doubles, but don’t make it that obvious. Just so that it gives your vocals and the meaning of what is being said in certain parts more power. Doubling is more common in rap music. If you are singing, rather than rapping, be careful when doubling because it can make your vocals too artificial and too pop-ish. On the other hand, if you are going for that pop sound, doubling might be a great tool for you!
In the chorus, you can record two vocal tracks and pan them between 30 to 60 – one to the left, one to the right. Another option would be to record a third track, which is placed in the center, but not as loud as the lead vocals in your verses.
Some people record one lead track and double it (copy and paste it) and edit them differently (EQ, compressor, pitch, etc.) This can be another great tool to make your vocals sound different in certain parts of the song, just like the panning advice I mentioned above. Try it out and see how you like it.
See you in Part II!
1. Use an EQ
Almost every base drum consists of two important frequency bands. The bottom of a classic base drum ranges between 65 and 110 Hz, and the top ranges between 3 and 5 kHz (sometimes 8 kHz). The bottom and the bass are two of the major components that will give your song/beat more punch, but oftentimes they get in the way of each other. To avoid that, apply a low-cut to your base drum, somewhere between 30 and 60 Hz. The bottom is where every base drum should have its heart, so adjustments in this band have to be made with caution.
Many base drums serve no practical purpose between 120 and 350 Hz, so you can lower this band by applying a bell filter. The top can get in the way of other instruments (e.g. toms), so it can be useful to reduce it, but oftentimes the top makes a specific base drum sound unique. Experiment and adjust it to your needs. If you like to add a second or third base drum, that’s ok, just keep in mind that every base drum claims certain frequencies and needs room to develop.
Add a compressor to reduce unwanted peaks, but don’t try to force your base drum into the mix by compressing it too much. Play around with the attack time of your compressor. Sometimes, longer attack times can improve the base drum’s percussiveness and enhance its positioning within the rhythmic structure.
Pitching your base drum can help your base drum fit the rest of the drums, and the rest of the song/beat. It can also help to fit your snare drum, and vice versa. Base drum and snare drum are like the yin and yang of your drums. They have to be complementary! Keep in mind that pitching can change the whole “essence” of your base drum, so you might need to EQ afterwards.
First off, be careful about adding reverb to your base drum. Too much reverb will make your beat sound muddy. In fact, this applies to all instruments of your mix. A nice little reverb on your base drum is a good thing, though. It’s a great tool to make your base drum sound more connected to the rest of your beat and to make it feel more natural. A dry, reverb-less base drum isn’t a bad thing in and of itself, either. For example, hip hop often uses a more detached base drum. It makes the base drum sound punchier. Playing around with reverb on different drum elements can even make the beat groovier. You might want to try adding reverb on the snare, but not on the base drum, or vice versa. Or you can apply different reverbs on different drum elements.
One last thing: Every base drum is different. Consider the numbers as a point of reference, but always let your ears decide. Do what YOU like.