Surround sound is when we use multiple audio channels in speakers placed around the listener to allow sounds to be heard from different directions.
The standard way of naming surround sound formats is with numbers separated by periods, such as 7.1.4
Understanding the names:
The first number describes the amount of speakers placed around the listener. Most of the time these are placed at ear height for the listener.
The second number, the number after the first point, describes the Low Frequency Effects (LFE) channel. This gets played by a subwoofer. This channel generally is band-passed to only play frequencies between 3 and 120 Hz. This makes sure that the LFE only contains the bass information we want. A lot of what comes out of the LFE channel is not so much heard but felt. The LFE helps make big moments in a film feel huge by shaking the audience a bit.
Now, we understand that 5.1 means that there are five speakers surrounding the listener and one LFE track. But where are the five speakers placed?
In 5.1, there are three speakers in front of the listener: Left, Centre, and Right. In a cinema, you may notice that you can’t actually see these speakers. This is possible because the speakers are hidden behind the screen. This works great to make it sound like the audio is coming from the picture itself.
The remaining two speakers are the surround channels - Left Surround and Right Surround, commonly abbreviated to Ls and Rs. These are placed somewhat behind the listener, so that sound can surround them.
As you can see a 5.1 mix requires 6 speakers and thus 6 channels of audio.
5.1 is the simplest common setup for surround, which we can make more interesting by adding more speakers around the listener. 7.1 has the LCR speakers at the front as before, but now with two pairs of surround speakers: the side surrounds placed to the sides of the listener; and the rear surrounds, placed behind the listener. Common abbreviations are Lss and Rss for side surrounds and Lrs and Rrs for rear surrounds.
Making it 3D
So far, all these configurations have speakers at ear-level, and so the sound effectively is only on a 2D plane. To add a third dimension, speakers can be placed on the ceiling around the listener. 7.1.4, for example, has four ceiling speakers spaced in a square above and around the listener. We can call these the Top speakers (Lt and Rt) and the Top Rear speakers (Ltr and Rtr). With this, we can make mixes with elements that move around the listener and above them.
Dolby Atmos
Dolby Atmos is an object-based surround format. What this means is that we can assign audio to objects and move these around in a virtual space. Then, when the audio needs to be played back, the Dolby system computes which speakers that audio needs to go to based on the setup in that specific room in order to best represent the virtual object environment. This means that an Atmos mix could be played on a 5.1 set up or a 7.1.4 setup, or something with even more speakers.
However, you may have also encountered Dolby Atmos music being playable on AirPods with “Spatial Audio”. How is this possible?
Binaural
Dolby Atmos and Spatial Audio through headphones works by creating a binaural render of the audio. This is a render that emulates surround sound by using audio processing to trick your brain into perceiving sounds from different directions, even though there are only two speakers. To simplify, the virtual objects and speakers are processed in such a way as to mimic how our head shape and ear shapes would affect the sound if it was physically played by speakers in a room.
Binaural renders only work on a binaural playback system, that is, a setup where each ear has a dedicated speaker, and each ear cannot hear the sound coming from the opposite speaker. A typical monitor setup with two speakers is not binaural as the audio from both speakers combines inside of the room – your left ear can hear the right speaker. This prevents the illusion of the surround emulation from working properly.
Since everyone's ears and heads are different, it is impossible to make a binaural render that will work perfectly for everyone. Most binaural renders try to use an average shape, so that it works fairly well for most people. However, for best results, each listener's mix would be rendered to match their exact head and ear shape. Apple's spatial audio can use a phone camera to capture your ear and head shape providing better results than rendering off an average.
For more on the topic of binaural audio, looking into Head Related Transfer Function (HRTF) is a good place to start.
Comments