Head-Related Transfer Function: What Is It and Why Is It Important
I’ve said it before, and I will say it again. Audio enthusiasts LOVE complicated-sounding acronyms to explain our hobby. I, on the other hand, like to demystify our hobby and put it in terms and examples that are easily understood. To this, I give you Head-Related Transfer Function (HRTF). Are you shaking your head already? Don’t worry; it’s not as complicated as it sounds, and it’s something that you may need to understand.
Let’s Talk About Those Ears of Yours
Most people have two ears, one on each side of their head. The outside of your ear, the part that your 1st-grade teacher used to pull on to get your attention, is called the auricle. This generally faces in the same direction as our eyes (forward) and directs the outside sound down the auditory canal to the eardrum.
All of this you probably know from…you know…having ears.
You are also probably aware that you can often recognize the direction of a sound. You are better at doing this if the sound is coming from somewhere in front of you. But you can pretty much do it no matter where the sound is coming from.
Also, You Have a Head
That melon of yours plays a role in how you hear. To locate a sound’s source, your brain analyzes the differences in the timbre of the sound arriving at each ear. You can test this at home. Play some music through a speaker. Now, face 90 degrees away from the speaker so that one ear is closest to the speaker. Cover one ear at a time and notice how different the music sounds. This is the effect of the sound having to go around your head to reach your other ear.
Sound in 360 Degrees
We (or brains really) are so good at noting the differences in what we hear in each ear that we can often locate the origins of sounds without thinking about it. The slight differences in the sound from ear to ear allow our brain to discern where the sound came from.
But it isn’t perfect.
Our ears face forward. This means we are better at locating sounds that originate in front of us. On top of that, we’ve all had the experience where we thought a sound came from one direction, but found out, usually when we faced that direction, that we were wrong. This is usually because the sound bounced off something nearby and “tricked” our ear (brain really). Enter Head-Related Transfer Function.
So What Is Head-Related Transfer Function?
In the most general terms, Head-Related Transfer Function or HRTF is a way of processing sound so that it more closely mimics the differences we’d normally experience. Rather than having two speakers put out the same sound in phase and allowing our brain to locate the sound between the two speakers (this is how stereo works), a sound designer can slightly alter the sound in one or both speakers. This can change how we locate the sound between the two speakers or even push the sound closer or farther away from us (perceptually at least).
With the advent of object-based audio and spatial audio, research into how we hear is becoming more and more important. It informs sound designers how to use Head-Related Transfer Function to better place sounds in the room with us.
Why Should The AV Enthusiast Care?
We live in a time where Dolby Atmos and DTS:X are the big things in audio. They claim to transport you into the middle of the action by moving the sound around you in three-dimensional space. Without HRTF, none of this would be possible or matter.
But wait, there’s more! What about those fancy Atmos modules that bounce sound off the ceilings and, with DSP, create the sensation of overhead sound? You guessed it! Head-Related Transfer Function plays a considerable part in this. By manipulating the timing of the sound, and its volume and phase, these modules can trick our ear into thinking that sounds come from overhead, or to the sides, without having a speaker there.
We even now have headphones that boast the ability to mimic spatial audio by playing with the intensity and timing of sounds. While not as convincing as having multiple drivers within the ear cup, some headphones have developed pretty impressive sound.
Feeling Like You Are Really There
Not only does Head-Related Transfer Function make it so that sounds are placed more convincingly in 3D space, but that they sound more real. If you really wanted the best possible sound (and you had Bezos or Musk money), you’d have your head scanned. You’d then have a model of your head and shoulders created with microphones in the place where your eardrums would be. Lastly, you’d have the music you wanted to hear recorded with that head actually in the audience (or room). This would literally give you the most realistic (and personal) audio experience possible.
Now, they’d have to model not only your head and shoulders, but also your ear shape and canal. It would have to be exactly the same as your actual head and ears. But, if you did this, you could create an audio experience that was identical to the one you’d have (with some exceptions) to actually being in the audience. You wouldn’t feel any of the low bass, but, other than that, it would be the same.
That is the point of a Head-Related Transfer Function. It is not only to try to make things sound like they are coming from a place they are not, but also to make them more realistic. While they have to do this using DSP and assumptions about how people’s ears and hearing work, they are getting better and better. And, for us, that is a good thing.
Head-Related Transfer Function – Making Sound Better!
Head-Related Transfer Function is essential to our enjoyment of audio in general. It allows us to perceive a sound in a space where no source exists. While Dolby Atmos modules are still iffy, with refinements of the technology and increases in processor power, we may eventually see a product that can produce 3D sound without installing physical speakers.
With more and more sound designers becoming aware of how we hear and using Head-Related Transfer Function to inform how they mix sound, we are getting more and more realistic movies and music. And if that isn’t the goal of HiFi, I don’t know what is.