THD 83 IDUN Brain-Sensing Earbuds that measuring neural activity EEG
foreign [Music] with another episode of THD podcast today we have a company joining us from Switzerland called edun and they're doing Electro and silificgram technology or EEG and yes I did practice saying that word earlier so we're gonna find out about this technology but without delay and it's it's basically they're embedding another sensor into earbuds and stuff so that's why we're going to talk about it um but without delay let's not forget about our sponsor the ulti association they're going to be at the high-end show in Munich and also there'll be uh cooperating with uh just before infocom in June in Orlando they'll be having another event there so you put the information right here on the screen and you guys can go check it out on your own the delay let's say hello Simon's joining us from Japan with some nice pink headphones yep Pretty in Pink and Mark melnikovich how are you doing wonderful thank you very much great to be here okay so lead product designer from E Dunn and uh yeah so doing EEG technology to put largely in earbuds I guess is that the core Focus or are we going into other spaces as well no exactly we're really looking at how we can take earbuds in-ear devices to the next level of understanding humans and that means understanding the brain so we're at a you know we're and we're at a very interesting point where hearables or earbud devices that you say are really transitioning from being just devices that deliver sound and media to eventually being things that are actually driving applications with you know the rise of multi-core processor technology uh the ear Os or com OS systems from companies like sonical and we want to be this layer which then is essentially this glue which helps understand human intention needs and also how the brain is responding to different media it's a sensor which we we really can't measure the type of signal for brain activity in any other way except by really going to the source um and it's really a beautiful complementary technology to the integration of different audio features AI to enhance audio experiences or the sound landscape around the person plus having motion so for example having a lot of the functions that you currently have in your Smartwatch but actually putting it into the earbud and so if Eden yeah okay so maybe just to illustrate this we might walk people through a presentation good thank you uh in Technologies it's really a company which is creating a new brain analytics platform for hearables now many times when we talk about hearables we don't we usually think about sleep but sleep is one of the most critical problems that we have in our in our daily lives and also throughout the world and understanding sleep as a metric and as an entity for for being a person is one of the things that we're currently applying this technology to which sounds a bit strange you know when you think about hearables but when we talk about sleep we talk about the evolution of where where earbuds are going in the future it's really interesting to look at the industry So currently if you're looking at companies like LG if you're at CS this year you hopefully have seen the LG Breeze earbuds these are earbuds that are also measuring the EEG or Electro esophagy from the ear it's actually their electrodes are a little bit outside the ear canal and then you also see great products coming from companies like Bose where you're using musical intervention to actually help the Sleep journey of people and what's missing is really understanding the feedback between what's happening in a person's brain what they really need and then how we can change the acoustic environment for them and you can see these are like very small highly packaged devices and that's a big challenge that we have at Eden how can we bring complex uh neurotechnology into the earbud form factor now for Sleep many people also then make the analogy to wearables so things like smart watches uh track sleep and they do track sleep but they track it in a way that's not as accurate as we get from being in a sleep Laboratory so an Apple Watch a pixel watch you're gonna have things like a heart rate sensor and a motion sensor which is giving you an indication for where you are in the in the Sleep Cycle but it's not really getting the information directly from your brain and also why that's super important in the future so traditionally if you want to measure brain activity you're going to have to put on electrodes to the top of your head preferably glue them there is dry electrode technology to make it easier but it's still a rather cumbersome device it's not a device which you're actually going to use day to day and at Eden we're really focused on creating this concept of the everyday neurotechnology device the everyday brain computer interface so we actually started from the electrode material from the earbud and then we use that as a platform to measure brain activity and that allows us to make things that are very integrated into the consumer lifestyle like earbuds and we also have already done validation of our technology so if you compare this full scalp technology to our in-ear system you're actually able to measure the brain signal the brain activity in a very similar way so if you think about how your brain actually changes during sleep you have things like changes in the signal of your brain the way it's oscillating you you'll have bursts of Elf activity you'll have changes to deep sleep and that's stuff that you can't really measure when you look at just the the macro architecture of sleep that you get from a smart watch and we also see this stuff can actually be connected up with how people are hearing or experiencing the world and part of what we do at Eden is building that technology but then also building how to understand how to interpret that signal from the brain so being able to extract signal patterns for things like the K complex or the Sleep spindle it's a very simple but ultimately very powerful algorithm because if we can do this throughout a night and also throughout the day we can really start to understand how you're reacting to different situations and then bring interventions such as Acoustics to help you focus better help you relax better help you sleep better we're really looking at bringing this a layer to earbud Technologies and just just for the the layperson the sleep spindles and K complex what what are those I mean I've heard of RAM and some of the others but so you can imagine that when you when you're awake your brain is at a certain higher state When You Close Your Eyes it starts to go into what we call the alpha phase so you're going from let's say 30 40 Hertz down to 10 Hertz of your brain activity and that's really what we measure so the electrical bio potential at the ear or on the on the surface of the scalp and then as you start to fall into actual sleep your your brain waves are going to shift another another little bit so go from about 10 Hertz to about 8 Hertz or six Hertz and then as you start to go into dreaming States you know we hear these these Concepts such as uh uh REM you know the random eye movement that happens when you're dreaming um these are all markers that we use in laboratory to understand where you are in the sleep cycle and usually you have to sit there and you look at this on a computer screen and a technician will sleep stage an entire night and we're really looking at ways to take algorithms and make this automatic and easy to use so that this complex history of how your brain is changing during the night can be easily interpreted through this mobile device and then used to build interventions like getting the the perfect type of music for you to fall asleep right if I like Rammstein and you like what what is your favorite music what do you use to fall asleep White House deep house white white noise usually yeah so like some people it's uh they're gonna fall asleep better or they'll have their optimal sleep Journey we want to help design that and understand it sorry Mark you may be going to come to this but um what we're looking at is an electrical signal where does that actually come from you mentioned bio uh potential I think so you can imagine that in your brain you have uh electrons and then you also have neurons right so a little neuron is going to be you know very small little thing that you're not really going to see but when they all get together and they start firing they start creating a an electrical potential and this electrical potential can be sensed on the surface of the scale and in this way we can basically get an indication and an understanding of what's Happening inside your brain okay it's just curious that you take this brain activity electrical into essentially a single you know one-dimensional uh potential voltage is It ultimately yeah exactly so yeah you can have many electrodes on your scalp you know some people use up to I think 1024 so you can have a very high density and you can learn a lot about the brain but then it's something which is never going to get outside the laboratory and what we're doing is bringing that down to just an ear tip just an ear tip electrode to measure stuff throughout the day so you can think about it as like a smart watch for your brain okay thanks yeah my pleasure and you can imagine if we can understand a bit of what's happening in the brain over a long time period you know let's go from days to weeks to months to years to decades we can actually start to see how your needs are changing throughout your life we can understand how your cognition how your focus is changing and one of the the higher purpose goals is to see if we can connect this up also with disorders that are happening in the brain you know ultimately dementia Alzheimer's and as you know also music and dementia treatment these are starting to be paired together the idea of music as medicine is also coming to the Forefront and this is one idea that we can connect up what's happening in the brain with the musical intervention that's coming through earbuds and then we can create essentially a new way for people to to live their lives as they're aging um and when we think about our product we first want to focus on sleep because there's like a big need in the market but there's so many ways that we can expand Beyond sleep into the daytime activities as well so things like measuring Focus as well as measuring interactions like the the direction of your gaze or things we can actually extract just from the ear canal so we are building a platform in addition to our earbud Technologies we just released our Guardian earbuds onto the market it's a B2B product so the companies and Enterprise can use this technology they can understand how to integrate it into their into their businesses how they can integrate into their own reference designs or really interested in connecting with more companies in the hearing space in the earbud space to see how we can build and integrate these reference designs eventually into consumer products on the market so beyond just a physical device we have our brain analytics platform so we have the ability to stream brain data through a mobile device or computers onto our Cloud platform and this can be done in a variety of ways we can go directly through our web page we use web Bluetooth connect devices to earbuds we also have a python SDK so companies can create more advanced scenarios like playing a type of music showing a type of media understanding what happens in the brain during that time and understanding then how we could do things like personalization of Music algorithms and our application layer within our Cloud platform allows us to actually learn different neuromarkers deploy different narrow markers and create a situation where we can have a really seamless experience between collecting brain activity data to building all these new use cases and then the challenge is bringing that down to an earbud form factor and this is where I think all these cool Technologies coming out like multi-core processors are really going to bring hopefully a paradigm shift in what an earbud is and going just from what people consider to be a relatively passive device to one of which is really integrated into their Lifestyles and is also replacing some functions of their smartphone and other Technologies so if this platform uh going Beyond sleep we're looking at how we can do things like acoustic attention with audio thinking about gaming how can we do more hands-free game interaction or having games that actually adapt themselves to our mental States right people get bored of games because they don't have the right balance between frustration of solving a hard problem and the flow of the game environment so in the future especially and this is the cool thing with generative media right when we have generative AI creating environments for users and the metaverse and also when we have um acoustic media which is being generated directly for users because this is you know going to be a next logical step we see we see this happening with visual media like still images then comes to video and now also the ability to adapt or actually create new musical pieces is also going to be coming and if we can combine that with our neurotechnology at the air it's going to bring some really interesting very seamless autonomous new experiences and also interesting is that when we read brain data one part is understanding brain state but we can also understand quite a bit about what's happening with the eye and the face so we're starting to develop these unique classifiers where you can actually do gaze detection and monitor how the eyes are moving in addition to doing the brain activity so you can imagine that when we start to think about integrating more UI or more ux into into the the earbud the ability to change the direction of how your 3D audio is being delivered to your brain is one thing that we can start to hook up with our in-ear Technologies and we were recently at the Mobile World Congress as a guest of Oppo at their Innovation event where we could show this technology to people coming from all over the world um and it was really cool and really engaging to see how people are reacting to it it's still a relatively large device that we're selling this is really to prove the you know the business case and also the application and we're really looking for a partner sales to help us take our reference design and package it into into a smaller system because integrating the ability to stream brain data with tws Technologies it's something that takes time to develop and it's something where we really want to be closer to the the device manufacturers to really realize these new products but a lot of our core technology it goes back to materials actually so one of our key USPS is the ability to measure this electrical brain activity from inside the ear canal and we do this by developing our own ear tip materials that are conductive and then we build our product around that and so doing things like tuning the Acoustics of the design tuning you know the the the way that sound is transmitting through the the um the sound channel in our ear in our earbuds there is also some of the interesting challenges that we have and then people are always asking ah but do you have a an interaction between the ear and sound is there a problem with measuring brain waves at the same time as measuring or delivering Acoustics and the answer is no we we have really good performance um so we've basically proved that we can measure the brain in a mobile way in an earbud form factor and now we want to basically take it to the next step connect with more people in the industry and basically I see how we're going to build the headphone 3.0 as people call it and the next evolution of earbuds and hearing and this really is a great segue into like our road map as a company we started really as a biosensor company we then prove that we could measure in ear EEG with our new device we're really connecting with all the business cases POC projects and in consumer or in B2B use cases and then we have a concept for our Guardian audio Eden audio where everything is packaged in a very small device first based on two WS technology but we're also interested to see what's happening with Le audio and what's really going to make the most sense in the future for combining the you know very easy to develop scalable audio platforms together with highly scalable brain analytics so we want to make the brain analytics easy to integrate into the audio platforms so we can get to the stage where instead of buying a tws chip you're buying it essentially in a tws slash EEG chip where we have our amplifier combined with our reference design for the for the ear tip and earpiece electrode system there's a Grand Vision for the company and we're very open for having conversations about the future getting feedback and seeing how we can all build this future of the earbuds together uh so um uh what I'm saying there's a couple of components to it there is the sensor part and it sounds to me like there's two there's two elements to that one would be the ear tip itself where you're getting physical contact and then there must be another like a uh an amplifier or some man handling of that input and then uh the second part of it would be somehow you process that data into so we've been looking at waveforms but you're obviously doing something more in terms of analyzing those waveforms to create results from a century and then another part of the package is uh what you do with that data how you interpret it which would be done generally outside of the earbud is that fair to say yeah I mean it's really interesting question because right now we stream all of our brain data to the internet and we do everything on our Cloud platform for extracting features and then having an API so that API can talk to your Spotify playlist for example I'm also seeing the future where there's going to be this hybrid solution where we have our Edge processing device which could be one of the new multi-core processors coming out is providing the ability to potentially You Know modulate audio at the same time as take in and process the brain signals and then a lot you know maybe 40 50 of the processing is happening on the earbud and then we still have that this longitudinal view for what's happening in your life how are your music preferences changing with your brain activity how's your sleep changing and having that really be more in the cloud so it's personalized and then also of course transferable over all the different Digital Services through API integration yep nice okay and uh just on a kind of a legal aspect is there any issue about the collecting those brain waves just as a as in collecting other data like because I can imagine your partners are potentially going to be people like Spotify that curate the playlist and Amazon music and they're going to be very interested to know what makes Mark happy what makes Simon mad and and so and then so yeah that's an interesting uh kind of dynamic as we collect more and more data um where could this end up so we actually have a narrow ethics board as part of the Eden company and our CTO mortex he has created our neuroethics Charter so when we're talking with B2B companies um we actually go through a process of them to understand what they actually want to do with this data and be sure that things are handled in an ethical way so that we're not for example stealing your thoughts or trying to steal the thoughts of people and of course keeping the user in control of who they're sharing their brain activity with um also when we talk about extracting these classifiers or neuromarkers from the data this is also another layer of protection so as we're controlling this so full value chain from collecting data to having an actionable Insight then that can also be restricted so that doesn't mean that you know just anyone would connect to an API endpoint and be able to access raw brain data so we have these protection mechanisms in place and it's also very interesting to see what's happening in different parts of the world with the interpretation of brain data because we have you can also extract tons of information from the smartwatch um the ability to like really characterize human behavior and then connect it up with identity is going to be I mean it's a pressing problem over pretty much all wearables especially as there's more data collected and stored so our goal is really to have the user in control um have a process for neuroethics that's also done together with industry representatives from neurotechnology and Neuroscience and really forming I hope a standard for how we can uh sustainably and also ethically build these types of neurotechnology consumer products so the current state is uh but this proof of concept is it what you would call it a proof of concept design we've definitely gone beyond the proof of concept so in our roadmap we so our proof of consult was essentially the Guardian development kit as we call it which was released last year and this looked more like a prototype product where it didn't really have a lot of scalability built into it so our current product we have great scalability of the actual ear tip material it's injection moldable it's transferable our system for producing Hardware it's also a much better supply chain we had a lot of compounding issues which was blocking you know just the general ability to to scale the product right now but it's definitely beyond the proof of console phase that we had earlier and now it's a product which is basically ready to scale uh we also have the new product development of a company cerebra Health in Canada where we're co-developing a sleep device because for the consumer Market in North America and in Europe so the product is very quickly going towards a masculability yeah and so for somebody who wants to get their hands on it wants to implement it the model would be that people would license the technology or buy components or both so right now for our Guardian earbuds it's essentially a subscription basis so companies can come we discuss and essentially rent the device on a monthly basis we're working on a reference design business model for the OEM for the later business business case where we can have a reference design license sell it and then scale it and that would include essentially the the way that we measure EEG so the ear tip production material and and as well things like firmware the type of potential amplifier to use and then how to integrate it into earbuds so we're looking at these two different business cases for scaling okay so I guess um it's uh be a fairly complicated process to implement at the stage well I don't know I think you know we're talking with some uh some of the larger companies about how to build reference designs so it's not that complicated I would say it's really about probably changing a few different processes so for example having conductive strips on the earpiece itself so we can conduct to the conducted ear tip and then having a more standardized way for integrating system on a chip modules with the hardware and this is also one of the interesting challenges for the future you know will it make more sense to do a die stacking chip ourself or work with other companies and do more of a system on module approach to the design okay so um it would uh this the device the current uh model Guardian earbuds um it has a fairly powerful processor in it or is it something that would be uh the software would be able to be implemented into existing uh tws chips yeah so we're actually working on a tws reference design in parallel it should be ready probably beginning of a Q2 so actually quite soon and there we're actually integrating with a Qualcomm chip so we have uh EEG streaming directly into the Qualcomm chip it helps with our partners and that's going to be our main reference design for doing earbuds at scale so it's actually almost ready and I'd love to talk with companies about yeah working with it and adopting it yep can I ask a little bit to you because that's what kind of processing actually is done so you're sampling uh So currently we sample uh 250 Hertz for the brain signal and we're not doing too much processing on the device itself because we wanted to bring for the development kit we want to bring a lot of this data into the cloud for doing algorithm development if we're going towards a specific use case where for example we're just measuring a change in Focus these are things that we can more easily do on the device and where it could make more sense to do it on the device for a specific consumer product so we're looking more at the let's say a microcontroller level processing and not needing an eight core multi-core processor right now to to do the product that we need shows almost like two kind of things one where you would collect an enormous amount of data and offline process then another one where you're essentially doing a a sample block by sample block uh characterization or looking for some characteristic or feature right exactly so if we have a narrow marker such as for focus and we want to just deploy that one mirror marker on device I mean that would be feasible um but for our development kit where many of the customers are still looking for you know the application the killer application that they want to develop for the technology and therefore we make a much more flexible platform in the cloud so I can pull in all this data you can run experiments much more seamlessly and then do the the data science on it yep yep very good yeah so what do you think about this type of idea of integrating brain activity with uh hearables uh I like it a lot and um all of these type of features um going into the earbuds so heart rate uh you know pulse pulse related but this one sounds a little more interesting actually oh thank you yeah and you know really I see we're going to be integrating all that stuff in the future you know either um and our own reference design if we take it to you know the third or fourth version or if we do it together with a commercial partner because of course I'm sure you know there are many devices already out there which are doing heart rate um measuring heart rate plus motion not too difficult and then when you add in our technology to that you have this great combination of understanding what's happening in the heart the brain and the body and I think when we have that we're going to have a really nice scalable powerful system for doing all these things like also measuring stress right because then you can measure heart rate variability then you combine it with your focus you combine it with you know where you're taking a walk in the morning were you taking a nap in the afternoon after lunch uh so I think it's like really an amazing exciting space yeah and um well and when you're the developer of this you can see this enormous range of possibilities but I guess from uh and implementer's point of view you might like to choose I really only want I'm going to make for example Sports earphones so this one's specific uh capability that I want to give to my users I don't care about sleep for example maybe they do you get my point you want to really try and consolidate onto one specific thing that you get a very robust uh response out of yeah and one of the reasons that we started with sleep is that we saw a lot of value that we could bring to the Sleep market so the Sleep Tech market and for this reason and also for the component Supply we had at the time we built this type of device which was something that is has sleep in mind it's not completely optimized for sleep so it can also be used during the day as a more normal earbud device um so solving sleep with some of our first commercial Partners made a lot of sense from the market and when we talk to other companies who are in the earphone space this concept of like okay what would be the application for neurotech earbuds it's still a bit open um and it's really great to see companies like LG coming into the market with these types of devices because it's first proving that there's an appetite from from the industry to bring these new types of Technologies but it's also going to be great for testing the market and understanding how people uh respond to that type of use case so they also saw the value in going with sleep first and in optimizing their their value proposition around sleep yeah but beyond that you know once you solve sleep where you might have earbuds falling out you have to have good contact people are moving around it's being used you know for probably six to eight hours already then transitioning to those daytime applications like tracking Focus tracking music that's going to be a lot easier and much easier to scale and then we want to have you know the market proven having devices in the market for asleep and then then going into these other um yeah amazing use cases is I think really going to be the way we go as a company uh that one do you mind if I ask this uh concept of focus or concentration let's say uh is there is there a metric for that I mean is there a way to say your concentration is at a level so right now companies are all doing it individually you know it's you know it's going to be the same as I think with many of these different applications that you you basically see a change a synchronization in brain waves um and doing it together with music so there's this company come uh concept of Flow State where you know your brain is very focused you're doing one thing could happen in sports it can happen in you know creativity and creative Industries so seeing how the brain is you know essentially shifting from one state to another is one of the things that we do with the device and having a real let's say biomarker for focus in the ear and being able to scale that is one of the big things are going to be bringing to the market we're doing that right now with sleep where you're essentially seeing like I said these shift in brainwave frequencies to define the the Sleep State yeah and it's kind of something similar to that for for folks if you were interested in a study headphone or earphones that can essentially prompt you when you're losing focus because you have the experience too or when you're working as well uh you can fall into a state where you're just not concentrating and just drifting and it could an hour could go by wouldn't it be nice if the iPhone just told you just go and take a break and come back and you know regain your focus rather than leaving you to drift off well or having your relaxation music or audio start to play automatically right so I really love this idea of the um the work time cycle you know where especially if you're doing like a new problem your your thoughts are like expanding and then Contracting and you need that time like your brain your brain plasticity needs that time to think about stuff and then go away from it and then you have memory consolidation and there's also a lot of cool um research happening in places like MIT where they're actually looking at memory consolidation during sleep for assisting learning so having things like having an audio queue when you're learning something during the day and then bringing that audio cue while you're sleeping this is one of the resource topics where they're trying to see if we can actually improve learning processes during the day by understanding and doing interventions during the night at the right sleep stage so I mean it's still you know really in the research phase but the potential is so amazing to really affect and improve all these different like small parts of our lives yeah all right so that's that's some good stuff very interesting stuff Mark um thank you thank you for joining us today um we'll put some information down below for people to uh to follow up and find out more information on edun and uh yeah so everybody like subscribe share all that good stuff and uh we'll see you on the next episode see you for now cool thank you sir thank you David
2023-04-16 16:45
