OIKLA's Detailed Technology Explanation
Firstly, it is necessary for us to make a clear distinction about terms used to explain the Problem and the Solution.
Sound Level or Volume
These terms will be used to describe the physical sound pressure level usually measured in decibel (dB) or pascal produced by a transducer (loudspeaker or headphones). This commonly refers to the “volume”, which we all change when the sound level needs to be quieter or louder on our sound systems.
Sound perception is used to describe how the brain “hears” the sound. Unlike propagated sound through a transducer, such as loudspeakers or headphones, there is no linear correlation between the sound measured in the real world and the spectral balance perceived by our brains - meaning that the frequency spectrum measured is not what the brain is perceiving. This phenomenon was first discovered and studied by Fletcher and Munson in 1933. Their study produced what is now widely known as the “Equal Loudness Curves”, which illustrate the phenomenon of human hearing.
Equal Loudness Curves: When the quality of the listening experience is affected by the sound levels.
Situation: Listening to audio through your loudspeakers or headphones
In this situation as the actual sound level changes, the perceived loudness our brain registers will change at a different rate, depending on the frequency.
Here is what this means:
At low listening volumes – mid-range frequencies, at which the human voice ranges: sound is more prominent, while low and high-frequency ranges seem to fall into the background.
The low and high frequencies are more prominent at high listening volumes, while the mid-range seems comparatively softer.
Yet, in reality, the overall tonal balance of the sound played back by the sound system remains the same, no matter what the volume - it is our brain that hears the difference!
Below is a visual example to illustrate how the brain perceives something different from reality.
Which one of these two lines is longer?
You may perceive the bottom line to be longer than the top line, but they are the same length!
Similarly the brain perceives sound differently than can be measured. Low and high frequencies appear to be quieter than the mid-range even though the respective sound level is the same.
Given the same sound level of each frequency range, mid-frequencies are always perceived louder than lower and higher range frequencies. However, the perceived difference between the low or high-frequency range and the mid-frequency range diminishes as the sound level increases, meaning that low and high are progressively heard louder as we increase the sound level.
The table below compares the relationship between three sound levels (safe, risky, and dangerous) and their relative frequency range perception using the same optical illusion.
Even if the sound levels are the same for each frequency range, the low and high-frequency ranges are the only ones that are perceived differently relative to the medium frequency range across the three different sound levels.
As sound levels increase, its relative perception does too: perceived from quieter (shorter) to louder (longer).
What are the problems that our technology solves?
The phenomenon discovered by Fletcher and Munson presents a problem for both the production (sound creation) and reproduction (audio playback) stages wherever there is audio i.e. Film, Music, video games, VR etc… .
One of the most fundamental characteristics of a good sound mix is achieving the ideal balance of frequencies to the listener (here gamer) during the production stages.
Another consideration to be taken into account is sound localization. In adults, sound localization accuracy depends on the frequency content of the sound heard. Since the sound level affects the perceived frequency spectrum, it can reduce localization accuracy for predominantly rich sounds, such as explosions, in low and high frequencies. At safer sound levels, our hearing significantly lacks low and high-frequency perception.
This begs the question: how are sound creators supposed to achieve a good and spatialized mix when the perceived balance of frequencies changes as the volume changes? And even if all sound creators agreed to create and mix audio at a single and standardized sound level, it would still not guarantee that the end listener (the gamer) would choose to listen to the same sound level as the creators did.
While sound creators, especially game designers and exec producers, cannot guarantee the sound they carefully crafted will be perceived as intended by gamers who arbitrarily change sound levels; gamers’ sound experience and knowledge of the provenance of sound (crucial in gaming notably in the Esports sector) is lost given that sound levels can be changed arbitrarily and consequently modify the original sound experience created for them.
Noise-Induced Hearing Loss
The higher the sound level to which we are exposed, the shorter the time it takes to cause hearing loss. In fact, from 85 dB, the time we can safely be exposed to halves for every 3 dB increase. Meaning that a steady sound level of 85 dB will cause hearing loss after 8 hours, whereas a sound level of 115 dB will cause hearing loss after only 30 seconds.
As we explored earlier, the higher the volume, the more our brain perceives the low and high-frequency ranges compared to the mid-frequency, which means louder sound levels enable us to access a much fuller and more exciting sound experience.
Gamers and, more generally, any listener is more often than not left to increase the sound level to obtain a more immersive and engaging gaming experience. The higher the sound level, the better we hear low and high frequencies. However, the higher sound makes them forget that even short, loud exposure (>1 minute) causes permanent hearing loss and/or a chronic condition like tinnitus.
The WHO estimates that 1 in 7 or 1.1 billion people worldwide are at risk of permanent hearing loss due to loud sound exposure in recreational settings.
In summary, different sound levels affect the overall quality of the listening experience in a nonlinear fashion.
Here are the biggest differences:
Higher sound levels offer a more thrilling and immersive sonic experience but are responsible for irreversible and premature hearing loss.
Lower sound levels are safer but do not offer the thrill of higher sound volumes.
What if sound could offer the richness and excitement of high volumes but at safe listening levels?
OIKLA’s solution offers just this! makes it possible elegantly and effectively.
Building a worldwide sound perception standard that enables sound creators to create sonic experiences translates into superior listening environments independently from the end gamer’s sound level.
How does OIKLA empower creators?
OIKLA’s technology allows sound creators to embed sonic information about any sound element into the form of metadata that specify the sound perception the end-listeners will be listening to regardless of their sound level. Thus, providing an alternative to the link made between the sound level and the sound perception. This new tool provides sound creators the ability to create and customize sound experiences that translate exactly as intended in a way that is currently impossible.
Here are two examples, one with OIKLA and one without.
Jane is the sound creator.
Mike is the gamer.
Jane designs a Gunfire sound - A gunfire sound is likely to exceed 110 dB in real life.
To have a realistic sound perception, the Gunfire sound would need to be played back at 110 dB. However, 110 dB would not only be deafening for Mike (well above what is considered safe), but very few sound systems could reproduce such high sound levels without significantly distorting the original audio.
Jane creates a Gunfire sound and selects a sound perception of 110 dB. OIKLA shapes the spectral balance of the audio to realistically represent the gunfire sound perception at a safer sound level.
When Jane is happy with the sound perception assigned to the gunfire sound, he embeds the metadata into Unreal Engine. The metadata specifies that the gunfire sound must provide the listener with a perceived sound equivalent to 110 dB.
During playtime, Mike sets a safe sound level (>83 dB).
OIKLA’s technology is embedded within the game itself through Unreal Engine and reads the gunfire’s metadata at “110 dB” and detects Mike’s current sound level at “75 dB”.
Once OIKLA is provided with these two pieces of information, it adjusts the gunfire’s spectral balance to be perceived by Mike as if it were played at a sound level of 110 dB. Note that OIKLA processes the audio in real-time based on the current gamer’s sound level.
The outcome is:
Mike’s hearing is safe.
Mike’s sonic experience was not sacrificed due to the low listening level.
The sonic experience that Jane crafted for Mike is preserved regardless of the in-game sound levels.
Why is it important to know the real-time gamer’s sound level?
Given how the different sound levels greatly affect the overall listening experience. OIKLA must know the sound level to which the gamer is exposed to at any given time otherwise it can’t process audio correctly.. Detecting the real-time user’s sound level (sound creator or gamer) allows OIKLA to determine the user’s spectral perception.
The user’s spectral perception describes how much of each frequency (20 Hz to then 20 kHz) the listener’s brain is “hearing” or better “perceiving”, which is not what can be recorded and measured using a microphone.
Given the exact sound comparison between the frequency spectrum of the recorded audio signal from a microphone and the user’s perceived frequency, we would see that:
In summary, knowing what the listener is truly hearing (the starting point “1”) allows OIKLA to understand the path it needs to take to get the desired sound perception without increasing the sound level (the arrival point “2”).
The two would always show different results
These differences would further differ as to the playback level (volume) changes.
How does OIKLA detect the user’s listening levels?
Depending on the type of sound system used (speakers or headphones), we offer solutions to reliably detect the gamer’s sound levels without constantly monitoring sound levels using a microphone.
Our intuitive smartphone app allows a one-time, intuitive calibration of the sound system to establish a relationship between the digital audio level and the sound level (sound pressure level) experienced by the gamer’s listening position.
A pre-calibrated list of headphones will be provided for headphones, thus requiring minimum action from the gamer. The calibration process consists of creating a list of headphones that OIKLA will test in-house in advance to identify their properties such as frequency response and the maximum loudness produced.
By performing this calibration OIKLA will create a profile for each headphone model so that the final gamer has, only to choose the headset for use. For wired headsets, the gamer will need to choose the model used from a dropdown menu. However, for Bluetooth headphones, the model will be automatically detected.
A Simplified Explanation of OIKLA’s Smart DSP Algorithms
According to the gamer’s sound level and the sound perception level specified by the sound creator in the form of the metadata, OIKLA computes and processes the audio through smart DSP algorithms based on psychoacoustic principles, which shape the audio frequency spectrum to achieve the specified sound perception, resulting in an enhanced sonic experience without the need for changing sound levels.
OIKLA enables a new layer of sound customization/enhancement for sound creators.
Our technology enables the emulation of how the gamer perceives a specific sound element within a game. There are two ways in which this emulation can occur.
One of the two is to assign sound perception similar to the one you would experience in real life.
Therefore, if a firing shotgun produces 120 dB in real life, the sound creator can assign that sound perception level within Unreal Engine to recreate a realistic sound perception thanks to OIKLA.
However, suppose the game developer intends to create a sonic experience that differs from real ones. In that case, OIKLA allows that by letting the game developers apply Hyper-Realistic soundscapes, which are not similar to how sound is perceived in real life! This ultimately provides the game developer community with a new layer of sound customization and enhancement that further augment the gaming experience.
OIKLA’s Sound Perception Standard (conclusion)
In conclusion, OIKLA’s sound perception standard enables sound creators to confidently and arbitrarily decide how each audio element in a video game will be perceived by the gamer regardless of the sound level during playtime.
This allows both ends (sound creators and gamers) to experience an enhanced and immersive sonic experience at safe sound levels. We believe that this will greatly expand the native audio capabilities of Unreal Engine while pushing the next-gen of sound in gaming and VR.