Interview
- Tatsuya Haraguchi Teiji Yutaka Makiko Kan
- Corporate Distinguished Engineer
The Future Expansion of "Kando" (Emotion) Through
Advancements in Technology
Corporate Distinguished Engineers are a team of certified technologists within the Sony Group who formulate and execute technology strategies while identifying signals of change, and support the development of talent in order to ensure Sony's sustainable growth. We spoke to three Corporate Distinguished Engineers who are specialized in technology areas related to entertainment - Tatsuya Haraguchi, Teiji Yutaka, and Makiko Kan - about what they think entertainment and creation will be like a decade from now.
—— What do you think the relationship between the physical and the virtual will be like a decade from now? What issues do you think will need to be addressed?
Haraguchi: In the world of entertainment, actions are already being taken to try to break down the boundaries between the physical and the virtual. To the point about a decade from now, I think people will not have to mind about boundaries in the same way as we do today. But no matter how much the world will change, it won't change the fact that the essence of entertainment lies in people.
Kan: With all this attention on technology, such as the integration of the physical with the virtual, the "people" aspect tends to get lost - so it's important that we keep the focus on that. Recently, when an artist livestreamed their concert, they said they felt isolated, not being able to feel the reactions from the audience. I think they were talking not about the excitement that would fill the venue, but about how difficult it had become to create stories between the artist and the audience members. In an actual concert venue, the artist is able to see the individual faces that make up the audience and think to themselves, "They look like having a good time," or "Oh, they brought their kids," while the audience is able to feel a sense of sympathy for the journey of life of the artist performing in front of them. This exchange of feelings is very important, and moving forward, the challenge will be how best these can be obtained in the form of information.
Yutaka: I think we need to reaffirm the advantages of both the physical and the virtual experiences. For instance, with physical experiences, the number of people who can come together in the same place is limited, but virtual experiences can bring together hundreds of millions of people from all over the world. On the other hand, the reality is that physical and perception-based actions like enjoying gourmet food are still incredibly difficult to recreate virtually - though this may change once we start seeing brain-machine interfaces that can stimulate the brain and sensory organs directly in practical use. I think in envisioning the future a decade from now, we will need to think through the ways to integrate the physical and virtual to provide the best Kando (emotional) experiences.
—— Can physical Kando experiences be recreated in virtual spaces?
Kan: I think that is precisely the challenge we must take on. Earlier, I mentioned the relationship between the artist and the audience, but I think there are things that are shared among the audience as well. That, too, is something that we need to decode. Figuring this out, for instance being able to develop a technology that enables audience to feel that they are sharing the moment with hundreds of millions of other people, will allow us to create new Kando experiences that are distinctly virtual, not possible in the physical world.
Haraguchi: I think the key is for the experience, even if it is virtual, to convey a sense of reality in a way that resonates emotionally. If you were watching a virtual live concert, for instance, it would be difficult to give you that sense of reality if you were just there watching. But if you were able to feel that you were sharing that moment with everyone else, or that your actions are influencing the overall experience, I think that would give you a sense of reality. As such, I think the key would be to create an experience of reality distinct to live concerts, in which the artist and every audience exist as individuals, and are able to effect change on the overall phenomena through their actions.
Yutaka: A clue to answering this question may be the Japanese idiom, "hada de kanjiru (feel with one's own skin)." Perhaps a technology that allows us to feel things in the virtual world directly (or, with our own skin) can lead us to create new experiences. In fact, I feel like we'd be able to create such experiences if we could develop a sensing technology that could capture everything that surrounds a person in one place, and reproduce the captured information in the form of sensory stimulation for people.
Kan: Coming up with that kind of sensing technology seems difficult. I feel like there is a lot of information that we perceive other than just our eyes and ears, like the body temperature of the artist and each member of the audience, the temperature and humidity inside of the venue, the vibrations of the sound, and so much more. But I do think it is a very interesting subject.
—— What are your thoughts on the idea of "real-time," a term that has become increasingly important with the expansion of the virtual?
Yutaka: The true thrill of real-time experiences is not knowing what will happen next. It is important to connect your actions' effects to the following phenomenon immediately. In that sense, I think if we can create a virtual world with zero latency (communication delays), that allows us to connect in such a way that we are able to feel the effects of each individual's actions, it will feel genuinely "real-time."
Haraguchi: This is almost a delusional idea, but I believe that it is possible to create an experience that is the same as real-time, even if it is not really real-time, as long as we can create a situation where people can feel that is actually happening right now. As long as you feel that you are participating in an event, and that your actions are somehow affecting it, the experience doesn't have to be in real-time. For instance, the three of us are experiencing a concert at different points in time. As long as each of us can feel that our cheering for the artist enlivens the concert and makes it more exciting for everyone, I think that will be real-time and reality for all of us. To make something truly real-time, a lot of people would need to experience it at the same time. But think about it in this sense, and it doesn't have to even be in the same moment. Instead, it could even be something that transcends the concept of time.
—— That's interesting. So would you say that a recreation of a past concert in a virtual space could be a real-time experience if people are able to access it and feel like they are there experiencing it?
Haraguchi: Yes. And I think this because of the changes we've seen in people's entertainment experiences. Not even very long ago, we would all sit in front of the TV to watch a show, but nowadays, there are more opportunities for us to watch videos on video sharing platforms on our own time. The difference in time has already occurred in individuals, and the sense of togetherness is now lost. I think if we could compensate for this with technology, we might be able to feel that we are sharing the same moment, even if we all watch the content at different times. In a world like that, we might be able to participate in new Kando experiences that allow us, for example, to feel as if we were actually at a concert that took place decades ago.
Yutaka: This sort of experience, an experience that distorts reality, may be possible if we're able to create a world with interactivity, where the artist at the concert or the audience members around you respond in some way to the actions you take. My area of work, gaming, is ahead of the pack in some ways when it comes to this idea. Games are a form of interactive content programmed to change in real-time according to the player's actions or decisions. All players can experience the same sort of excitement no matter when they play the game.
Kan: In order to create these kinds of virtual experiences, we'll need a level of quality that really makes people feel by heart that they are there. At the same time, I also think it's important that we keep in mind the value of the physical. Even if everything in the physical world could be replicated by the virtual, I don't really feel it would have the exact same value.
The physical is valuable in that they are the places that actually exist, or experiences that occur in reality. I think a direction to combine the physical with the virtual to enhance the entertainment experience would be interesting to explore.
Yutaka: It's true that most of the information contained within a person is learned from the physical world, so having something that is linked to this information recreated in the virtual world would make it more valuable. But because we've each learned different things from the physical world, it will be important to show these things in a way that is optimized for each individual.
—— The next question is about AI. How do you think AI technologies will contribute to the future of entertainment?
Haraguchi: AI has begun to be implemented in music production in order to support creativity, but I think its role will become increasingly important in the future. There is heated debate as to whether there is inherent value in a work created by AI. But I personally expect AI to be able to enhance artists' creativity by making it more convenient for them to engage in creative efforts, or by making the production process more efficient. I also predict that in the future, we'll see AI evolve in tandem with the endless pursuit of people's curiosity, and that we might see, for instance, the emergence of AI that creates new and unexpected kinds of music.
Yutaka: Images in games are created in real time through calculations, so the processing power of the game console affects the quality of the images. To create realistic movements, physical simulations are used to calculate the movement of objects. I believe that by using AI, the amount of calculations can be greatly reduced by predicting the results of the simulation without having to calculate to the end. In other words, by making it possible to process a large number of simulations, for example, each piece of a large building that collapses can be calculated, resulting in more realistic expressions. In this way, AI makes it possible to create expressions that were previously impossible, and I believe it will give creators new creative freedom.
Kan: The key will be how deftly we can utilize AI to make the work of creators seem "handmade." Creators must be the stars, and the AI is there to optimize their creations. The basic principle when it comes to AI is to set a certain evaluation parameter, after which the AI will continue learning to try to increase the value of that parameter. When human values change, the evaluation parameters must also change. What we think is good now might not be considered good in five years, and it is the nature of human beings to be constantly in flux in terms of how we think and feel - meaning what is good and what is bad changes from moment to moment. In the world of entertainment especially, there are no correct answers, and we'll need to continue updating the parameters themselves, in accordance with how creators sense these changes in how people feel, the minute changes in their emotions.
—— What are your thoughts on sensing technology in the future?
Kan: What we want to do is to reevaluate the process of sensing. Sensing as it is now has become an increasingly data volume oriented. Usually, it collects as large amount of input data as possible at the initial phase, and only uses the necessary data at a later time. This does not look sustainable, and I personally think people are tired of being surrounded by enormous volume of data. I believe we can further sophisticate the performance of sensing by acquiring only the necessary data at the phase of input, and processing only the bare minimum amount of data necessary.
—— What are your thoughts on the current interest in the terms "transmedia" and the "immersiveness" of content?
Haraguchi: In my understanding, transmedia is the idea of going beyond the boundaries of conventional media, such as motion pictures or TV, to create more moving stories, but I personally would like to view the idea from a broader perspective. I think transmedia could transcend not just the boundaries of different forms of media, but also genres such as music and video, and even the five senses - for instance, smell and taste - and that it is important to continue exploring unprecedented Kando.
Kan: I think we'll gradually see the boundaries between professionals and amateurs, creators and fans, disappear. We at Sony have been making efforts towards this end, such as organizing an experimental event where people with no experience can play a musical instrument through the power of technology.
Yutaka: I think Sony's sound technologies, such as 360 Reality Audio, will be useful when it comes to creating immersive experiences. For example, in a game, a highly immersive experience can be created by changing where the sound is heard based on the player's position, or by making the sound muffled if there is a wall. We want to provide creators with technology like 360 Reality Audio to help them create these unprecedented immersive experiences.
—— Finally, can you tell us about your vision and efforts when it comes to content creation moving forward?
Yutaka: In recent years, one of our major themes has been "Maximizing individual Kando." The process does not end when we deliver the content - the goal is to move users' hearts, and we need to consider the whole process, including how the content will be received. Of course, different people have different tastes, and will even have varying preferences based on their emotions in that moment. What we want to do is to research and develop a new kind of experience in which the content changes dynamically according to how the user feels in that moment, instead of staying fixed as it was when it was released.
Haraguchi: This is more about one's mindset, but I think what we need to do is to build a new world of creation, while taking on some risk. Our vision does not end with just developing technologies, but in using these technologies to produce new creative experiences and businesses. To do so, we will want to figure out how best we can use these technologies proactively, while protecting the rights of artists and creators.
Kan: Within the Sony Group, there are many seeds of technologies that have not yet been implemented in society, but I feel that recent improvements in computing power are making their practical use more feasible. We hope to help deepen entertainment by providing creators with these kinds of technologies.
Profile
Tatsuya Haraguchi
Engaged in the development of information recommendation technology as a software engineer after joining Sony Corporation (current Sony Group Corporation). Since, he has worked in a variety of areas in Sony Music Entertainment (Japan), from management of social networking service specialized in music , management of web media, development of mobile apps, and the management of music streaming business. His current work involves exploring advanced technologies in the entertainment business; applying advanced technologies to various forms of entertainment, such as music, anime, and gaming; and promoting efforts to incubate new businesses leveraging technologies.
Teiji Yutaka
Was involved in the development of the first PlayStation®, and has worked in the gaming field for a long time. Currently leading research, including mid- to long-term research, on advanced technologies that could contribute to the advancement of interactive entertainment in areas such as gaming. Engaged in research on a wide range of advanced technologies, including realistic CGI based on real-time ray tracing, and UI technologies such as emotion recognition using intelligent processing.
Makiko Kan
Ph. D in Mathematics. After joining Sony Corporation (current Sony Group Corporation), she worked in the research and development of error-correcting codes, particularly algebraic codes such as Reed-Solomon and BCH, and probabilistic codes such as Turbo and LDPC. Made proposals regarding LDPC codes and related technologies for digital broadcasting systems worldwide, has been awarded many standard-essential patents, and was in charge of proposing formulas and implementing them in receivers. Also involved in the research and development of deep neural networks and the technology used to implement them - sensor signal processing technology - as derivatives of the Bayesian networks used for LDPC decoding.