00:00:00:

00:00:00: you.

00:00:00: Hey there, my name is Usheen Lonnie and welcome to Audio Talks presented to you by Harman.

00:00:05: Now this is a podcast where we explore the overlap between technology and creativity every time and today we are joined by Dr.

00:00:14: Mathias Johansson who is the Vice President of the Cognitive Systems Lab over at Harman and this is a team pioneering the intersection of AI and immersive audio.

00:00:24: So we are going to talk about how artificial intelligence is redefining what it means to listen, how it's making sound more responsive, more emotional, and more personal than ever before.

00:00:35: Now, from gaming to automotive to live music experience, we're going to explore how audio is evolving from something you control to something that understands you.

00:00:45: Welcome to the podcast.

00:00:46: Matthias, it's great to have you here.

00:00:47: Thank you so much, Oshin.

00:00:49: It's a great pleasure to be here, and especially to talk about these topics, the intersection of immersive audio, audio in general, and AI.

00:00:58: We're very passionate about that at Hormone and me in particular, so really excited to be here.

00:01:02: Fantastic.

00:01:03: That is actually event diagram, immersive and AI and audio and technology.

00:01:07: At the middle, you'll see Matthias Johansson.

00:01:09: So thank you so much for being here.

00:01:10: You're the perfect guest to take us through this.

00:01:12: Now, let's start at the beginning, because immersive audio has really become a buzzword in many sectors, such as gaming, such as entertainment, and particularly these days in automotive.

00:01:23: But I think it would be helpful for myself and for our viewers and listeners just to get a definition from somebody who's actually working in it.

00:01:30: What does immersive audio mean to you?

00:01:33: So I think that's a truly great question because the industry is a little bit vague sometimes around this, but it really like started, you know, we had stereo music that we had surround and then immersive became this name for three D audio essentially.

00:01:46: But to me, it's it goes beyond just three D audio.

00:01:48: It's about realism.

00:01:50: It's about presence.

00:01:51: It's about.

00:01:52: not feeling like you're listening through headphones or speakers.

00:01:56: It's more like the sound is truly coming from around you.

00:01:59: You're really surrounded first in sounds.

00:02:01: It's really about engagement and presence.

00:02:04: Fantastic, yes.

00:02:05: That totally resonates.

00:02:06: From personal experience, I know that, say, if you're in an immersive sound environment and it's all kind of working beautifully, and then if something doesn't quite track, it's like the illusion.

00:02:17: just breaks.

00:02:18: How important is it to be accurate?

00:02:21: It's super important.

00:02:21: It's fundamental.

00:02:22: When you think about how we as human beings, we're not used to listening to stereo format, to speakers in front of you.

00:02:29: Everything you hear around you is really immersive in reality, so to speak.

00:02:33: And gaming, for example, I'm having a pair of gaming headsets all.

00:02:37: When you're gaming, And if something in the sound is off from the visual, same thing goes for a video or a movie, right?

00:02:44: But if the audio is off, then the whole sense of realism disappears.

00:02:50: And talking about gaming, it's so fundamental to have proper spatial audio and precise spatial audio.

00:02:58: Because there, it's not just a layer of atmosphere.

00:03:00: It's really about the whole, it's about survival in some games, right?

00:03:03: It's about performance.

00:03:05: because you can locate a certain sound where an opponent is really quickly using your ears if you have precise spatial audio and you simply cannot do that if you just have regular stereo audio.

00:03:16: Right,

00:03:16: yes indeed.

00:03:17: So like if you're kind of walking through a post-apocalyptic landscape as sometimes we like to do in very good video games and you hear a sound kind of behind you or above you, that can make the difference between, you know, virtual life and death.

00:03:29: But talk to us a bit about perceptual modeling because I don't quite understand this and I'd love it if you could explain it to us.

00:03:36: So it all starts with perceptual modeling.

00:03:38: Perceptual modeling means how do we as humans perceive sound?

00:03:42: So when a sound comes from a certain direction, what is it that happens?

00:03:45: Well, you know, we have two ears and we can model the path from the source to our ears through perceptual modeling.

00:03:53: It's not just how a microphone listens.

00:03:55: It's very different from how our brains actually listen to sound.

00:03:58: and we decompose the sound into the direct wave.

00:04:02: and then the early reflections and the reverberation the ambience.

00:04:06: these three elements are really key in our perception of spatial of localization.

00:04:12: so we model this in different ways to capture like really realistic immersive sound.

00:04:19: the early reflections are very important to capture distance.

00:04:22: And there, we use what we call mirror sources to model this with added diffusion and so on.

00:04:27: Whereas when it comes to the reverberation, it's more about the atmosphere.

00:04:31: Am I in a large space or in a small space?

00:04:33: These are what we call macro level statistical properties that we need to model very accurately there.

00:04:39: It's about spectral properties, use of technical terms and correlations and these kinds of things that we are mathematically modeling.

00:04:45: And then of course, to bring that into say, headphones like these, you know, you need to, this is pretty heavy maths that are involved there, and to cram that into a headset in a headphone is pretty difficult.

00:04:57: And that's why you look at a lot of spatial audio implementations today, they're not always that realistic.

00:05:03: And one of the things is simple, that this modeling step is pretty hard to say the least to get right.

00:05:09: And, and then to cram it into this with low latency, because it needs to be super low latency, as soon as you have it, you know, The synchronization is off.

00:05:18: What happens then?

00:05:19: Well, the illusion is lost.

00:05:20: So that's completely fundamental.

00:05:22: And that's like this fundamental aspect of perceptual modeling that is so important.

00:05:27: That's phenomenal.

00:05:27: Thank you.

00:05:28: Well, I understand perceptual modeling so much more now than I did a couple of minutes ago.

00:05:33: Thank you very much for that, Matthias.

00:05:34: But it can't be easy to make all of this stuff happen.

00:05:37: Talk to us a bit about how you make it so.

00:05:40: I mean, do you have stations of engineers and people around the place?

00:05:45: What's going on under the hood?

00:05:47: I mean, the way we attack these problems, you know, Harman has research labs in many different locations because we're spanning a professional audio, consumer audio, and spatial is a big part of that as well as automotive, all of these areas.

00:06:00: So we have a research in spatial and immersive audio in all of these, the different application areas.

00:06:06: And we recently started a new lab in Copenhagen that's focused on headphones audio for consumer headphones, gaming headphones and so on.

00:06:14: And There's a long process here, like the acoustic pre-development process for a specific model is very intense.

00:06:22: Like six to nine months, we have a set of more than forty golden ears, as we call them, that are involved in testing and giving feedback to the developers on the audio.

00:06:32: And then in the end, we also involve real end users, gamers, if we're talking about gaming headsets, for example, to really hear how is the average user Perceiving this and what's the feedback there so we can move on and improve the overall experience.

00:06:46: So it's really a long process here, but we're approaching it very systematically and our legacy is in engineering.

00:06:54: Engineering excellence, the science of how we perceive sound and that's where our roots are.

00:06:58: That's where we always start.

00:07:00: Fantastic.

00:07:01: And I'd just like to say a big hey to the team over in Copenhagen.

00:07:06: hung out with him at Copenhagen.

00:07:07: I've visited Soren and his amazing collection of gear and it's just such a pleasure.

00:07:11: you really have some of the best people there so that's really awesome.

00:07:14: but you mentioned there about the legacy of sound and this is of course it's sort of the north star of Harman and all of the brands like JBL, AKG etc.

00:07:23: You talked to us a bit about how this magnificent legacy that goes back decades and decades and decades how this kind of filters down and influences and informs what you do in immersive audio.

00:07:34: oh it's absolutely fundamental because you know everything we do.

00:07:37: like.

00:07:37: i said we have these different divisions and so on but they're all based on the scientific core and harmonious one of the scientific pioneers really how human beings perceive sound and that's that's sort of what we're really using in all of our different domains and to us it's all about precision audio right having the precise accurate reproduction of sound so that you get away from this feeling of listening to a box, no speaker box or headphone.

00:08:04: It's like this sense of being there.

00:08:07: And precision and accuracy is what allows you to have the emotional impact that we really need.

00:08:14: It's super fundamental.

00:08:16: And we have the experience from shaping the sound of some of the absolutely most demanding listening environments in the world from high-end luxury home theaters to live arenas.

00:08:29: and to precision tune vehicles.

00:08:31: like you know premium audio systems in cars today are really advanced is a fantastic sound systems some of the best that you can experience in fact and over to studio monitor zone.

00:08:41: so all that legacy is actually built into every product we're developing and that's really something that I think you know at Harmon everybody's so proud of and so passionate about.

00:08:51: it's this engineering excellence and getting that to become.

00:08:55: these experiences where you are present in the sound and really engaged with the music.

00:09:00: So the TLDR is all of these decades of experience and expertise and R&D and manufacturing and engineering expertise.

00:09:10: It all kind of distills down just to bring immersive sound to life.

00:09:13: It's, you know, taking it outside of the box and really being the best sound you can possibly experience on the planet.

00:09:20: A little while ago chatting with some of your colleagues and they were talking about the importance of the eclipsa open format.

00:09:28: I mean

00:09:28: how would you say the eclipsa fits into this picture?

00:09:31: I think

00:09:31: eclipsa is a great example of where immersive audio is heading.

00:09:35: It's going everywhere right and eclipsa for those who don't know it's an open royalty free immersive audio format.

00:09:43: So it really brings the power of immersive to endless numbers of people now, content creators, YouTubers, et cetera, because YouTube is supporting Eclipse.

00:09:54: A lot of devices are starting to support it.

00:09:56: This just shows that all of a sudden it's possible to create immersive audio, which five or ten years ago was more difficult.

00:10:02: Now the mechanisms are there, both the tools to create and to distribute, and then the ultimate you know experience the rendering of the sound.

00:10:11: because we also need to think when we talk about formats and so on.

00:10:14: somehow it can be a bit confusing i think to a lot of people when we talk about immersive audio.

00:10:18: immersive audio is partly the format.

00:10:22: the format is the carrier of the content but once you have an immersive format it's not enough.

00:10:28: you need to create an immersive experience.

00:10:30: that's what we talk about always right in the end it's not just the format but it's the experience and that then requires Deep expertise in how it's the last mile.

00:10:40: It's the speakers, the headphones, and the software that renders this, this immersive scene to you.

00:10:47: That's sort of very important.

00:10:49: And that's where we see a lot of exciting developments now going on in immersive audio.

00:10:53: We're working on it in many ways, like with spatial audio and also looking at perceptional modeling.

00:10:59: We already mentioned, but also how AI fits into this to personalize this immersive experience for you.

00:11:05: Right, and that's fundamental as well.

00:11:07: Fantastic.

00:11:08: Well, I'm very glad you mentioned everyone's favorite two-letter acronym, the A and the I. A flashing light just went off in the podcast studio because we're in prime podcast territory here.

00:11:17: Could you talk us through how AI fits into this picture of immersive audio?

00:11:22: Yeah, I mean, AI truly is the next big leap.

00:11:25: You know, traditionally you really had to manually adjust everything in your sound system.

00:11:30: They're essentially because you have a lot of different technical settings when you buy an advanced sound system, and they are pretty intimidating to most people to actually try to adjust and get to their liking.

00:11:41: With AI, the picture is changing completely.

00:11:44: The way we now look at the future, it's very exciting because instead of having us as users having to adapt to the product, and be a technical expert almost to read the manual and really understand what's going on and adjust the sound.

00:12:00: We've got to change that.

00:12:02: There's going to be a paradigm shift where products now start to understand you as a user.

00:12:07: And one of those examples is like I talk about changing the sound of your system.

00:12:13: You might say, okay, it's a great sounding system, but the music I'm listening to now is a little bit sharp for some reason.

00:12:19: It might be the source.

00:12:20: That is not perfect, right?

00:12:22: How do you adjust that?

00:12:22: For most people, just trying to go into a graphical EQ and trying to adjust that is pretty difficult.

00:12:28: You really need to be an expert and so on.

00:12:30: So what if you could instead have a virtual sound expert with every speaker or pair of headphones that you buy?

00:12:37: That's how we see now AI being able to make a kind of a revolution in how products adapt to you and how they personalize the sound to your context, your desires and needs.

00:12:49: And that's just fundamental, because with the power of large language models, we're able to understand what a person means.

00:12:58: When the user says, like I said, the sound is too sharp, for example, or I wanted to open up for, well, with our Harman expertise, we can also ingest that expert knowledge of what that statement actually means.

00:13:11: What does it mean if sound is sharp or open or warm in technical terms?

00:13:15: That's what we've built decade by decade, that knowledge base, right?

00:13:19: And now we can use that in those LLMs and have like really a virtual, personal sound engineer come with you, shape the sound as you want it on your terms, like in plain language, not technical language.

00:13:31: So, so AI is truly a game changer here at the next big leap.

00:13:36: It's not just about the spatial positioning of sound.

00:13:39: It's also how the sound itself becomes aware of you and your needs and really adapt to you.

00:13:44: That's

00:13:45: wild.

00:13:45: So, you know, There could be a scenario where I'm listening to a record at home and I say, hey, make it warmer and it will just make it so Star Trek style.

00:13:55: Is that what's going to happen?

00:13:56: Absolutely.

00:13:57: Yeah.

00:13:57: That's exactly what's going to happen.

00:13:58: You can really just describe how you want the sound to feel and then with the contextual understanding that we can build into these systems and with language models and these advanced what we call RAG models where you can ingest expert knowledge, we can really do that for you.

00:14:15: And we're doing a lot of data collection to ensure that the system really gets you and really understands what you mean and adapts to you.

00:14:23: And this is grounded in all these research that we've done and talked about here in our previous part of this talk here.

00:14:30: So it's a very exciting development in my mind because it makes sound less technical and less reactive and more adaptive to you and more human in a way.

00:14:41: I

00:14:41: love that.

00:14:42: I have four words for you.

00:14:43: Take my money already.

00:14:45: If you don't mind, that is just so up my street.

00:14:47: I can't tell you because often you're listening to a record and it's sort of, you know, not quite there.

00:14:52: And if you just say something to change it, to make it more your style, this sounds like music to my ears.

00:14:57: Excuse the pun.

00:14:58: But again, coming back to these, you know, decades of expertise and all the teams all over the world, all the folks in Copenhagen there, this can't be easy to make this happen.

00:15:07: Could you talk a bit about some of the challenges and some of the obstacles that maybe you've broken through to bring this to the world?

00:15:14: You know, there are lots of challenges.

00:15:16: It's always easy to have an idea, but the difficult part is to execute on it and make it happen in also a form factor that works, right?

00:15:23: We can't just build, you know, we build a lot of super expensive high end sound systems, but they're not for everybody.

00:15:28: And we really want to democratize great sound.

00:15:31: So that's one of the critical things.

00:15:33: How can we get like?

00:15:34: really?

00:15:35: It needs to be snappy.

00:15:36: When you talk to a system like that, you need to have the response really fast.

00:15:40: Latency is one of the most important performance indicators for us.

00:15:44: It needs to happen fast.

00:15:45: When you ask the system for something, it needs to react fast.

00:15:48: Of course, that's a challenge.

00:15:50: to get it through this.

00:15:51: Get it to work in these chipsets and make it fast because these are pretty powerful demanding models that you have to use, obviously, to get it right.

00:15:59: That's the other thing, get it right.

00:16:00: The answer needs to be right and it needs to do the right thing for you because there's no such thing as, we can't get it wrong, right?

00:16:08: It's very annoying if you have an automatic system that doesn't understand you.

00:16:12: So there's also a part here, we actually have in my team a number of psychophysiologists, psychologists, because there's a large element here, it's not audio, it's understanding of humans, how we react, how do we want the system to react so that it feels empathetic and intuitive and not intimidating as that, because there's there's that balance that you have to strike.

00:16:34: So it's technical challenges and there's challenges in how we understand psychology, human psychology.

00:16:40: And that's what we're expanding into.

00:16:42: And I think that's an unusual area for most audio companies, right?

00:16:45: But we have that background also in Harman from before, because we've worked with also with what we call Neurosensing in the automotive context, where we're able to sense how humans in cars, their heart rates and their eye gazes and what that means.

00:17:00: about their human state, about the mental state, the cognitive load and so on.

00:17:03: We're doing all of that in automotive, which probably not a lot of people know about, but that's a background we have.

00:17:10: and then combined with this psychoacoustic knowledge that we have, we have an excellent background for doing this.

00:17:17: And we're showing some pretty exciting results now in our labs.

00:17:21: And it's even a fun feature.

00:17:23: Let's imagine you have this virtual audio assistant.

00:17:26: Of course, the whole idea is to make it sound better, but you can also have fun with it and say, hey, why don't you make it sound like it's coming from the apartment next door, right?

00:17:35: you know it will be able to adjust it for you and make it like this sound instead and it's a fun feature as well.

00:17:40: so you know this is what we found when we started testing on users.

00:17:43: they would do this crazy stuff and they thought it was super fun and So, you know, that's also part of it.

00:17:49: This is whole user testing itself.

00:17:51: We get a lot of useful feedback.

00:17:52: That's

00:17:52: phenomenal.

00:17:53: I think having that rave next door setting would be so much from, particularly if you had people visiting, yeah, the neighbors, oh my goodness, what a nightmare.

00:18:01: That's just no end of pranks and fun happening there.

00:18:04: Brilliant.

00:18:04: But you kind of mentioned like how much of this technology and this expertise and this psychology is going into the cars themselves.

00:18:12: And at Gamescom earlier this year, I did take a note.

00:18:15: You said that the car is becoming one of the most exciting acoustic immersive environments available.

00:18:22: Could you talk to us a bit more about why you said that and what it means?

00:18:26: Yeah, I mean, cars these days, premium car audio systems from Harman, these branded systems are really amazing.

00:18:33: You often have more than twenty speakers in surrounding you.

00:18:37: It's a very controlled environment.

00:18:39: So it's an environment that lends itself very well for immersive audio and for really advanced.

00:18:44: immersive experiences, and we use it both for entertainment, obviously, for immersive audio.

00:18:49: We're also having more and more immersive audio formats going into the streaming services, so it's much more accessible than it used to be a couple of years ago.

00:18:57: So it's obviously for entertainment a huge opportunity.

00:19:00: This is where a lot of people will actually today experience the best immersive audio experiences, I would say, because you can have some really amazing sound systems in modern cars.

00:19:09: But that is not just about entertainment.

00:19:12: You know, this is another area where immersive is breaking, you know, out from just entertainment and movies.

00:19:18: It's also going into aspects that you might not have thought about like safety.

00:19:22: So when you're driving down the road and there's something coming up, I'm in Sweden, might be an elk running across the street or something.

00:19:31: And then our ears, our auditory system is the fastest way for us humans to react actually much faster than visual feedbacks.

00:19:38: There we can use immersive audio to create directional alerts so that I hear an alert from the exact location of where that elk is and I can take a break or do a maneuver, whatever is the safest action in that sense.

00:19:52: So it's also creating all this kind of new use cases that we might not think about first, but PD localization is a big thing.

00:20:00: We talked about gaming.

00:20:02: When cars become autonomous in the future, imagine the kind of gaming experiences you can have in the car then, right?

00:20:08: Or movie experiences.

00:20:10: And it's a very controlled environment.

00:20:12: So for us, you know, we're working across both software and hardware and signal processing and AI.

00:20:18: And we can adapt the sound to this environment because the environment, the room is very, very important for the overall experience.

00:20:26: And when we're able to control it and know exactly where you are, in relation to the speakers, we can tailor the sound so that you get an amazing experience.

00:20:34: So automotive is, for sure, like one of the most exciting areas of immersive audio today.

00:20:39: That's

00:20:39: phenomenal.

00:20:40: I can just see you kind of driving across a beautiful forest in Sweden and the alert saying Harry Good and it's, you know, an elk coming out of the side.

00:20:49: That is very, very useful stuff to be sure.

00:20:53: And yeah, I can only imagine autonomous cars gaming, network gaming, you know, movies.

00:20:58: It's going to be one of the like the prime entertainment spaces of the future, I think.

00:21:02: And yeah, like you say, it's important for the sound to be kind of ahead of that movement of humanity into this era of autonomous cars.

00:21:10: But anyway, I would love to just flip over, if I may, to another area where immersive sound might be important.

00:21:17: I'm not entirely sure.

00:21:18: I'd love to find out from yourself.

00:21:20: And I'm of course talking about the original immersive audio experience, the live gig.

00:21:24: So you're there with a ton of people and you've got a great, like maybe a JBL sound system.

00:21:29: You know, would you say that there is a place for immersive audio in the live music arena?

00:21:34: Pardon the pun.

00:21:34: Yeah,

00:21:34: absolutely.

00:21:36: Very much so, right?

00:21:37: And it's already happening, I would say.

00:21:39: You know, we have immersive concerts now are becoming more and more common.

00:21:43: So you already can experience.

00:21:44: I've had some experience that are just phenomenal because when you think about it, like in these big arenas, You just have the frontal stage.

00:21:51: You only have so much possibilities to create real artistic expression.

00:21:56: But when you have speakers around you, you can create sounds from everywhere.

00:21:59: You can create these magical experiences.

00:22:02: So this is already happening.

00:22:04: And in professional, we also have a company, Flux, that is doing some very exciting stuff on spatial rendering for professional use cases, like arenas and so on.

00:22:15: And it's super exciting.

00:22:16: And it's because the thing is, it's not just audio there.

00:22:19: you have the lighting effect.

00:22:21: So we're getting more multimodal experiences as well.

00:22:23: I would say both lighting and audio is already happening.

00:22:27: And then the third mode of interaction is of course haptics, tactile feedback, which is a thing for gaming.

00:22:33: But you know, I expect that this will also become a bigger thing in other areas, maybe not live arena so much.

00:22:38: But I talk about live arenas, that's one thing when we're there physically.

00:22:42: And that's already happening.

00:22:44: And it will be more and more common.

00:22:45: But the other thing is what we're envisioning for the future is virtual concerts or live streamed concerts where I can sit with my headphones in my home and feel like I'm really next to the artist.

00:22:58: For me, like I'm an old timer, I really like Elton John and who does that?

00:23:01: I mean, he's an amazing musician.

00:23:02: Just imagine being next to him at his piano in a small intimate setting with a few people around.

00:23:08: And there's like, technically, there are not that many barriers left.

00:23:11: There's some, you know, to really create that precision, both in visuals and audio.

00:23:16: But that's a possibility that there with head tracking and with precise perceptual modeling and spatial audio, spatial vision and so on, right?

00:23:25: You can have some fantastic concerts remotely in the future.

00:23:30: So I think, yeah, for live, it's absolutely happening already and we can expect much more to come as well.

00:23:37: Oh, that's amazing.

00:23:38: And another edition of the podcast, which is possibly coming very soon.

00:23:44: to our podcast feed.

00:23:45: We had a chat with Little Boots, aka Victoria Hesketh, who was one of the founding musicians for the ABBA voyage, the kind of live concert experience.

00:23:53: And that's really fascinating about the overlap between AI and VR and XR and live experiences.

00:23:58: So yeah, I think, like you're saying, immersive and AI and XR, it just opens up this whole new world of live concert experiences.

00:24:06: So Thank you for that insight.

00:24:08: That was brilliant.

00:24:09: But I'd love to come back to your Gamescom presentation from earlier this year, and you mentioned something about moving from fixed to fluid.

00:24:17: It sounded very cool.

00:24:18: Can you take us through that?

00:24:19: Yeah, it really means, in my mind, that the audio experiences become seamless across your life.

00:24:27: It's like, imagine listening to music while you're cooking at home, then with these AI-based systems.

00:24:36: also with the ecosystem approach that we're now building.

00:24:40: The system will know you're in the kitchen, it will adjust the balance for your movement and so on, and then you sit down for dinner with the family, it'll lower the energy, soften the sound, redistribute the audio, and then later you will maybe go over to the living room, start a movie, and then it can automatically, based on the context, change at its senses, reconfigure this into a cinematic mode.

00:25:03: And this can all happen automatically in the background, with possibility, of course, to say, no, I don't want to do that.

00:25:10: You don't need to go into, because today, if you have a multi-speaker setup or a multi-room setup, that can be pretty complex to navigate and to set up for different use cases, for background music, for movie, and so on.

00:25:22: But now, we're seeing a future where this can all happen automatically, seamlessly, based on detecting the context of your situation, where you're in.

00:25:33: and you can just adapt.

00:25:34: Again, this is harnessing the power of AI and reasoning models to understand what's going on based on very limited input, what the system knows about the situation, and then orchestrate this experience to have the best possible sound for you in the context you're in.

00:25:51: So this whole idea is around.

00:25:54: What we want to do is with all the different speakers and devices we're selling, say you have a sound bar, a few Bluetooth speakers, we want them to work together as a cohesive unit so that they bring out a better experience than just individual or disconnected items.

00:26:08: It's really about connecting things and making less fixed and more fluid as you move across the different uses of your day of audio that it just happens in the background, follows you, go over to the car, it follows you as well.

00:26:23: So I think that's a big shift we're seeing as well.

00:26:26: in the future.

00:26:27: about audio in general.

00:26:29: Wow, yeah.

00:26:30: And this kind of builds on a lot of what I understand from the world of Harman and Bluetooth with AuraCast.

00:26:37: We had a previous podcast recording with Mikael Herrier and Chuck Sabin from the Bluetooth SIG talking about how these different devices in the Harman ecosystem can connect effortlessly, seamlessly with AuraCast.

00:26:50: And now this is like the next level.

00:26:51: It's kind of orchestrating the experiences from an adaptive viewpoint from an emotional viewpoint, you know, it sounds from what you're saying that this system sort of knows the rhythm of your day and will actually, you know, put you in the best possible context to have a nice experience with your audio.

00:27:10: Is that fair to say?

00:27:11: Yeah,

00:27:11: exactly.

00:27:12: I mean, it's also what makes AI so exciting for audio because it really transforms technology from something you operate to something that's just in the background that understands you.

00:27:22: And so that that's the future we're trying to build here sound that feels intuitive and effortless, like you said, and very human, right?

00:27:29: That's that's Wow.

00:27:31: Wow.

00:27:31: So from what you're saying, it's kind of moving out of just sound for entertainment or sound for listening to music.

00:27:37: It's sort of like this immersive audio environment, like that is the backdrop to your to your everyday life.

00:27:44: Is that right?

00:27:44: Absolutely.

00:27:46: And that's exactly the goal, right, to make sound really adaptive and responsive and emotionally intelligent.

00:27:51: So whether you're driving or working or working out, the whole idea is like your audio system should know how to support you in that context and in that moment so that you can just focus on whatever you're doing, right?

00:28:04: Not having to fiddle with technical gear.

00:28:06: Again, take my money already.

00:28:08: But you mentioned that Scenarios have been orchestrated in the background.

00:28:13: Is this where AI kind of overlaps to join all these different dots?

00:28:18: Yeah, so AI orchestration is really the backbone of this.

00:28:20: This allows devices to communicate and adjust and learn to adapt to your preferences and learn how to use these different devices together.

00:28:30: So AI orchestration is really the backbone of this whole thing.

00:28:34: Yeah, it's creating this adaptivity in the system.

00:28:36: So this is tech that really actually understands you because it analyzes what you like.

00:28:41: Yeah, but I mean, that's exactly it.

00:28:43: Right.

00:28:43: It's why we talk about immersive audio five or ten years ago.

00:28:46: It was about only about placing sound in space.

00:28:49: And even though that's also fundamental, right?

00:28:52: It's also going beyond that now.

00:28:54: It's really to make sound that is aware of you, of your movement, your mood even and context.

00:29:00: So again, it's like.

00:29:01: putting you making more human technology right or like intuitive.

00:29:06: that's that's the point

00:29:07: amazing my goodness and this ties in with what you mentioned earlier about the sensors in the car kind of detecting certain aspects of your your mood say if you're getting tired if you need a bit of help driving.

00:29:16: this is utterly fascinating but i'd like to invite you.

00:29:20: as i have you here i'm going to take advantage of your.

00:29:22: I'd like you to take out your crystal ball and talk to our viewers and listeners a bit about what excites you the most about the future of immersive audio.

00:29:30: Well I mean it must be that that we're talking about here and it's that sound is becoming truly intelligent and empathetic and also static and also moving beyond entertainment into sort of every aspect of our lives and into like I said safety and so on.

00:29:46: it's like it's becoming truly a big part of the life.

00:29:50: And underneath this is that whole trend of software-defined audio.

00:29:55: That is also something we're exploring here.

00:29:58: It's how we use both the advanced processing, the spatial audio processing with the hardware design to create those precise and engaging experiences.

00:30:07: And that's underlying all of this, and it's an absolute fundament.

00:30:11: But this AI-driven audio orchestration would turn devices into these connected ecosystems that are really responsive, context-aware, and forget about fiddling with all these menus and stuff.

00:30:24: These systems in the future will understand you and they will be able to adapt, just like you had a real expert, sound expert, with fifty years of Harman background that's built all our systems at your home and you just say, hey, I want to have it like this and then can you fix it?

00:30:41: And the system will be able to do that.

00:30:44: this is exciting.

00:30:44: And I think also when we talk about like starting with the spatial rendering itself, you know, this is today maybe a premium feature, but I see that becoming a baseline expectation because it becomes so much more engaging, so much more immersive, the sounds when people start to expect this, because the difference in the experience is so big.

00:31:03: So yeah, it's a lot of stuff that's really exciting as you can hear that I think is just on the verge of happening now.

00:31:10: And now at Harman, we're just proud and passionate about trying to lead that change to turn this cutting edge technology and innovations into experiences.

00:31:20: So, yeah, it's a lot to say here.

00:31:22: And I could go on forever.

00:31:23: But I think that that captures it.

00:31:25: Thanks, Mika.

00:31:26: That's absolutely amazing.

00:31:27: I just like thinking of those possibilities there, dialing up, you know, get me Raul Gonzalez, I want the front of house sound for Enrique Iglesias at the Hollywood Bowl or like, you know, all these different scenarios.

00:31:37: being able to Zap up an expert is just an amazing, amazing concept.

00:31:42: So thank you for everything that you and your colleagues are doing.

00:31:44: The future sounds better than ever.

00:31:47: But I have one more question for you.

00:31:49: And it is a question that we ask all of our VIP guests, everyone from Carl Cox to Raul Gonzalez to Carl.

00:31:55: AR, Raman too, all the good folks and that is to choose a song for our title playlist and so this could be something that just speaks to who you are and what you like or it can be something that sounds awesome and immersive or whatever you like.

00:32:09: Okay so what's your choice going to be Matias?

00:32:11: There's a lot to choose from but I would like to choose a fellow Swede, Salam Al-Fakir.

00:32:17: Keep on walking, just a Justin.

00:32:20: So beautiful

00:32:20: choice.

00:32:21: Absolutely.

00:32:22: Ten out of ten.

00:32:23: No notes.

00:32:23: That's great.

00:32:24: And I am going to myself contribute a little track to the playlist this time, which is by a band called Tomorrow Comes to Harvest, which is a kind of constellation being put together by techno genius Jeff Mills.

00:32:36: And they've got a track called The Happening.

00:32:38: And they recently did an immersive audio playback of this, I believe, over at A.D.E.

00:32:43: with some very good friends of ours.

00:32:45: And it's just brilliant.

00:32:46: It's transcendent music and it's transcendent creativity coming through in the immersive audio format.

00:32:53: So listen, Mathias Johansson, Vice President of the Cognitive Systems Lab at Harman.

00:32:57: Thank you so much for joining us.

00:32:59: It is super clear that you and your team are building something that goes far beyond software, hardware, and AI.

00:33:06: It is actually about creating the sound that feels alive.

00:33:09: So, two's intact, Mathias.

00:33:10: Thank you so much, Rachene.

00:33:12: Really appreciate it.

00:33:13: Thank you.

00:33:14: It's been awesome to chat.

00:33:15: Thank you.

00:33:16: OK, folks, the TLDR there.

00:33:18: If you take one thing away, it's at the next frontier of audio.

00:33:22: Isn't just spatial, it's intelligent.

00:33:24: AI is turning passive playback into adaptive experiences that respond to how we move, to what we feel, and gives us exactly what we need in the moment.

00:33:35: So if you enjoyed today's conversation, please subscribe wherever you get your favorite podcasts.

00:33:39: And join us next time as we explore more stories where sound technology and emotion meet in a very good place.

00:33:47: For more exclusive content, some behind-the-scenes goodies and maybe even some competitions, be sure to follow us over on the Instagram.

00:33:53: you can find us at Audio Talks podcast.

00:33:56: Thanks so much for joining us.

00:33:57: We will see you next time.