What's new with audio technologies? - Part I: speakers

The famous Consumer Electronic Shows (CES) has, for many years, been focused on display and audio technologies before becoming a major event for other technologies such as cell phones, automotive, VR and many more.

I'll discuss what's new in display and camera technologies in upcoming editions of this newsletter. Still, for now, I wanted to go over a few innovations happening in the audio world and share my viewpoint on the ones I believe are most promising with a focus on speaker/sound rendering techniques, microphone technologies and audio processing innovations.

Audio 101

Today, the vast majority of speakers use a magnet that pushes a membrane using electricity to create sound waves. The frequency and amplitude (~=force) by which the membrane vibrates allow generating sounds with varying pitches and intensities.

Human ears can "pick up" sound waves between ~20Hz and ~20kHz that is, we can hear a membrane vibrating between 20 and 20000 times per second, given a certain intensity and at a certain distance. The intensity vs distance ratio is quantified in decibels or "dB" - a metric that characterises the pressure that a sound wave exercises on a given surface. It is measured using a microphone and the amount of electricity generated by the microphone's membrane itself vibrating as it receives the sound wave emitted by a speaker.

Old gramophones could only reproduce sound between 250Hz and 5kHz (below human capabilities) whereas modern audio equipment can be easily designed to go beyond what the human ear can perceive.

Worth noting that the animal kingdom is often a lot more capable than us to perceive sound; for example, bats have a sound "pick up" range of 9kHZ to 200kHz which they use for echo-localization.

Echo localization means that their brains can process the sound they self-emit and receive (with our 2 ears, humans somewhat have that same "binaural" capacity - i.e., distinguish sound coming from the right or left but bats are much more precise at estimating distance and direction of sound waves than us). That's right, audio technologies aren't just about being emotionally moved by a song - there are also topics like localization or sound tuning which a lot of high-end audio systems offer when you set up a new home theater audio system.

The bat example is also a perfect summary of the 3 main components that need to be addressed when you discuss audio: sound actuation / sound perception / sound processing. Let's tackle each subdomain starting with sound rendering.

Part I - Speakers/sound rendering

Other techniques than "planar-magnetic-speakers" (a magnet pushing a membrane to create airwaves) already exist with few people knowing about them. Let's review some of them:

• "Bone-conduction speakers" are found in some headsets or hearing aids and designed to generate vibrating patterns that bypass the air/eardrum by their placement and direct contact behind the ear or cheekbone and resonation through human bones, notably the skull. The range covered is said to be in the 250Hz-4kHz range and also uses small magnets or sometimes MEMs (micro-electromechanical systems also used in recent microphone implementation) to create the vibrating force. So what this means is that your ear is NOT the only sensory system able to pick vibration patterns and interpret them. A famous story goes by the fact that classical composer Beethoven who ended up deaf was still able to perceive sound and compose using this technique. While the fidelity isn't as great as traditional planar-magnetic-speaker that can accommodate the 20Hz-20kHz human range, this technique has merit in low-power, portable applications such as VR headsets or sports headsets.
• "Electrostatic speakers" sometimes called "smart surfaces" use a thin, electrically charged diaphragm to produce sound. From a physics perspective, they are similar to conventional speakers except that the membrane (replacing the magnet in this case) and resonator layers are both electrified. Electrifying the diaphragm requires high voltage current and one of the main drawbacks is sensitivity to humidity, and the lack of bass reproduction due to the cancelling effect of ambient vibration given there's no resonating enclosure like in traditional speakers. The pros? They are thin, very thin.
• "Screen sound or acoustic sounds": Found in some cell phones, TV screens (like LG's Chrystal Sound OLED technology) use "piezoelectric" transducers placed behind flat screens to oscillate the entire display surface. Electricity is used to activate certain types of material (like crystals, and ceramics) that are naturally charged in electronics by nature. If I'm oversimplifying, these materials are kind of magnets. Typical tradeoffs include maximum sound intensity, reproducible frequency range, and volumetric requirements. It's no surprise that piezoelectric isn't widely used as cost, performance and reliability issues still need to be addressed. Yet, an interesting property of piezoelectric material is that it "combines" both speaker and microphone capabilities (that is uses electricity to generate vibration while vibration can be used to generate electricity), which can have many interesting implications in other use cases including haptic/touch sensing.
• "Directional speakers": one of my favourite topics is still at the research level/low-level adoption rate - unfortunately. Imagine the concept of "sound bubbles" that would isolate you from annoying surrounding conversations (the famous "cocktail party effect" that suggests that our brain can isolate certain sounds or conversations in a general "brouhaha" environment but that has clear limits). Excluding physical sound barrier isolation (like a dedicated sound room with enclosed walls), there are 3 ways people are trying to create sound bubbles in the open air: 1/ cancel ambient noise which your car is already doing but it mainly cancels out engine, road or wind noise but not human discussions that may be ongoing amongst passengers or with individual cellphone conversation 2/ use proximity speakers or headsets (like so-called smart "headrest" in cars) which tends to cancel ambient surround noise but isn't designed to mute or reduce one's voice against other that may be seeking privacy and not wearing these "close-proximity" speaker systems 3/ use complex, expensive speaker arrays - often paired with some head localization scheme to steer sound direction (see picture below). I'll address 1/ and 2/ in Part II of my newsletter on the audio topic. For approach 3/, the idea is really to funnel the sound waves in a way where they reach people's ears as opposed to spreading wide in the open air. Contrary to common belief, so-called "parametric sound arrays" that emerged in the late 90's, don't use a set of speakers physically oriented at different angles but rather non-audible ultrasound emitters that, through phase and amplification tricks, can "shape" the direction of a primary audio wave.

• For having personally tried one of these, the effect is quite remarkable - in a lab setup... Cons? It mainly works at close range, it doesn't like obstacles and the ultrasound steering technique used can affect animal earing (like dogs'). Plus the implementation is way more expensive to implement than traditional speakers so you'll see those mostly used in commercial applications (e.g., dispensing audio advertisement messages in a fixed direction in shopping malls or the case of underwater communications). So for more "consumer" oriented applications, techniques focused on 1/ or 2/ are currently favoured

Conclusion

It is interesting to see the parallelism between the physics of sound propagation and wireless transmission technologies (see my former article https://www.dhirubhai.net/posts/frantzlohier_activity-7155737197386637312-RVwx).

Audio technologies are still evolving and some techniques may or may not ever see mass adoption. When it comes to fidelity/quality reproduction of recorded sounds, it is often a matter of end-to-end system-level topics that need to be addressed with various tradeoffs depending on the use case.

For part II of this newsletter and on the topic of audio, I will cover microphone and audio processing technologies, notably as currently used in VR/gaming or the future of automotive active safety.

#audio #speaker #VR #MPEG-4 #MEMs #microphone #display #activesafety #audioprocessing

I hope you enjoyed this article. As always, feel free to contact me @ [email protected] if you have comments or questions about this article (I am open to providing consulting services). More at www.lohier.com and also my book. Here to subscribe to this newsletter and see former editions.