Skip to main content

tv   Press Here  NBC  July 7, 2019 9:00am-9:31am PDT

9:00 am
this will week, hollywood debates the future of entertainment, sit-down theater or stand up action. here hearing voices, how computers are learning how to understand what we say no matter how we say it. our forbes contributor, maribel lopez. this week on "press here." good morning i am scott mcgrew. there is a debate in the entertainment world over the future of story telling in movies. putting images on the screen for people to watch passively or is
9:01 am
it more interactive using virtual reality to bring the viewers in on the action. this is a new technology called spaces. al spin off of movie maker dream works. everyone who participates, sort of their own personalize termin terminator movie. it is very real and it is very enterta entertaining. >> our ceo of spaces. i presume he holds the high score, he's the world's leader expert of virtual reality and joined by our maribel lopez. so this thing is in the malls and i want to compare it to laser tag. i can't tell if that's going to make you angry or not because it is so much more. to somebody who never seen it, it is sort of laser tag.
9:02 am
>> yes, you describe it. one way but everyone describes it all kinds of different way. it is part laser tag. it is part theme park and movie and game, all meshed up together in its own kind of new things. the best way i like to say is you can go to movies and watch a movie and our experience, you step into the movie and you become the star. >> to those who have not done it and i did it. it is phenomenal. it is more than just the head set. you got sensors all over your body and when you look at your own hand. it is a robot hand. whatever theme you have to be doing which makes it convincing and heads up. >> it is a premium experience where your full total body emulsion so we track your hands and body and you are able to
9:03 am
walk freely in this world. if we are all in it together, we got to go do this, it would be so fun. we can see each other and reach out and giving each other high fives and interacting with each other. >> when i did this with one of your competitors, one of the things i was struck by was how heavy the equipment was. when do we get to do this in a way that allows you to move freely and natural. >> that's a great question. i think we have been at it for six years, three years of dream works and three years of spaces. the hardware that we use gets smaller or better and better quality and higher resolution and lighter or just more powerful. and so there is this kind of evolution of the hardware where we continue to upgrade yoour experiences with the latest technology. things are going all to one head sets and without needing towards
9:04 am
our backpacks and that is going to unlock entirety a new category of experiences. >> for decades, movies have been a passive experience, maybe of the exception of concerts and other live events, how big do you think this will be? is it going to be an opportunity for the coming years? >> like any new technology, it starts off-ish and stands. just like the internet and mobile computing. this is perceptional computing. we are bringing entertainment. can you imagine the future of let's say theme park. theme parks are mechanical and they're really fun and cool. but they're not necessarily interactive and they're not digital. i think that's what it is going to unlock. >> you give the analogy. it is something as we as
9:05 am
reporters are going to say everybody will have a 3-d printer. it is just around the corner and it is going to be phenomenal and never quite is. people experience what you have done and i think maybe they change their minds a little bit. although it will not be at-home experience for quite some time. >> we are billing something that you can't do at home. i like to think of our years of dream works and our management would say build the best experience and you can build a business around that. our aim is to deliver at least a location base and attraction and experiences. the best thing you can do so people get excited about what we are as a product and the technology. >> is the location based approach though ultimately going to be doomed by technology. we have seen home entertainment
9:06 am
signature pla planting a lot of the experience. >> or home videos. >> exactly. >> so how quickly is that going to happen and is that sort of a long-term threat or will there be a mark for peopet for peoples to go some place to experience something. >> there is a natural tendency for people wanting to leave their home, not necessarily for everyone. when we think about what we are building is an activity that a group of people can do together and that is kind of our focus. it is not necessarily building vr or this technology or that. let's create an experience that requires us to go out and have fun together. i think there is always going to be a place for that. and more importantly it is just an expansion of entertainment offerings out there. the home expansion that's going to be amazing that you can do at home. the other home is going to
9:07 am
continue to get better. i am thankful to hear that you love the experience, what we have planned for the next year and year after, we keep on trying to push the limits that's possible. >> you will change the theme. today is terminator and later is something else. >> it leads me to the next question. video games and this is sort of a video games tend of lend themselves into conflict and physicality and guns, i have no objections to guns in video games. is it a natural of well, we are going to shoot something. it would be nice to create a situation where a stereotyping an eight-year-old girl birthday party experience that may be does not involved killing robots. the technology and the physicality may make it difficult. can you imagine a vr situation that's more about ponies? >> well, i can imagine one about whales and penguins.
9:08 am
we actually have an experience with national geographics. >> i just set you up and i did not know it. go ahead and give it. >> well, we actually created a virtual reality show format where a real life national geographic explore can stand on stage and take the audience to antarctica or where ever it may be. we did one this other weekend where local from the bay area, e researchers and world experts with manta rays. >> do you see this integrating with other events that are happening such as e-sports. an opportunity taking somebody playing vr in your venue and have that interphase with the e-sports community? sth
9:09 am
>> that's a great question. we do see different types of locations emerging. you experience and you are playing together with people, that's cooperative. there is still a scoring element of what we have. people try to compete for a high score. what's nice after that is you get a video. people like to watch people doing vr. but, more importantly like to see the competing oing or worki together. that's a natural extension to your point. you are going to see that for sure. >> shiraz i am running up a commercial break. what's the thing in there that makes you go oh, that's so cool. i can't believe we are pulling that off. it is really exciting to hear people say i just didn't know that was possible. i didn't know it exists.
9:10 am
that was amazing and ex sill ra rawe have worked on this for yes in a warehouse in burbank. we are not sure how people are going to react. it is exciting to see that we are bringing people together in a fun way using technology in a way that they'll never forget. >> shiraz, the ceo of spaces, thank you for being with us this morning. >> up next, telling a really bad joke, we are trying to understand why computers don't understand us when "press here" continues.
9:11 am
9:12 am
welcome back to "press here," there is an old joke of the way programmers think. it goes like this. the wife tells her engineer husband please go to the store and buy a carton of milk and if they have eggs, get six. he comes back with six cartons of milk because they had eggs. the humor behind the joke if you can recall it still stands. it is hard for computers to understand what it is we mean. that's the challenge for programmers all over silicon valley. this is where they are teaching
9:13 am
computers to transcribe the language. you say it in a phone call and other computers can give you a record. sam lam, he's the guy behind the blue dot on google map as well. thank you for being with us right now. >> what is the hardest thing your computers have to figure out when hearing people talk? >> it is very difficult because there is a huge variation when people speak words. the different accent and the noise in the background. people have different pace and also the language itself is very indicative. a lot of words have similar
9:14 am
pronu pronunciati pronunciation. when you talk to doctor, they talk in different jargons and there are a lot of acronyms which they understand but you don't necessarily understand. new words are being invented everyday in silicon valley in the world. a lot of names that are not a common english names. there is a lot of immigrants here so all of this makes it difficult to understand the speaking language. >> yet, this has been kind of the for 25 years, it all seems like it is out of the reach. the thing that i wonder about is technology, the power behind computing has advance so much, why is it that this is always out of reach to have good natural language processing and understanding? >> there are two major reasons.
9:15 am
one is again it is complicated. >> a lot of things are complicated. >> this one got humans involved. >> true. think about it -- when people talk about ignition, the first thing people thought about these days are siri and alexa. it sounds like it already solves the problem. when you are talking to alexa, you are talking to a robot. it is a check box that listens to your short question and answer the question. but when you talk to another person, can other people in the meeting, it is ten times or more times complicated. you are talking to multiple people with different accents and different pace and distance
9:16 am
between the speakers and microphone. people interrupt each other and people talk much faster. they don't necessarily speak in a correct way phonetically. they talk about things that they discuss last week. they have a lot of contexts. if you don't have the context, you don't know what they're talking about. >> is there a particular accent where it is tougher than others? i am guessing scottish is one of them. would you say boy, we just need a team working on that one. >> accents are particularly difficult, right? even in the u.s., people in texas, boston or california, they talk differently. between the u.s. and the u.k. is different accent.
9:17 am
it is celebrated byseparated by. spelling is different is different between the u.s. and u.k. and now in silicon valley, we have a lot of engineers from india and china and other countries, south africa and austria. they all talking different accents. so the way people have to deal with is use a big data. for us, we have millions of hours of autoda data to train different models and learn all kinds of different accents. >> i want to pick up on the machine learning and big data along with alexa common which was very bounded. if we are going to move beyond the alexa style home model asking one question, what do you see where this technology can go?
9:18 am
we use machine learning and big data and better energy and we get there, what kind of things are going to happen? what kind of business models are going to emerge and what kind of applications will we see happening? >> this is pervasive. think of alexa, how many times do you talk about alexa? think about how much you talk to other people. we are talking right now with four people. when you go home, you are talking to your family and children and you have five or ten colleagues every week. people spend most of their time talking to other hum beings rather than talking to a robot. so the business model where cases with infamous. we see because if you think about it -- how many words do you speak in your entire life? there are studies have shown a
9:19 am
person may speak 800 million words in their lifetime. >> sam, i want to fit in one more question. your case for others of ai, i can upload a file and that's exactly what people said with little distractions or errors, rather. what is the legality of recording somebody to text, is it any different than the legality of recording somebody's audio. i can't record you on the telephone. i am not recording you, i am transcribing you, is there any legality change there? >> um, it is a good question. i guess there is some gray area there. recording the voice, there is a federal law that has to be, you have to be apart of the conversation to record, have a two party consent to record.
9:20 am
take notes, i don't think there is a law. >> i am not taking the notes but the computer is but somebody is taking notes. >> if you are taking notes, there should be no i mitation because if you can type really fast or you go to the doctor and actually there is a system who's typing the words to take notes. i am very practical. does this help you improve your communications and help you improve your productivity? >> right. >> the law is not ecstatic. >> the law never quite catches up to what technology is doing. >> sam liang is with us. i appreciate you being with us. >> thank you very much. >> up next, all that machine learning, we'll talk about the
9:21 am
news. when "press here" continues.
9:22 am
welcome back to "press here," our contributor has been in the forefront of digital news. he practically invented online newspaper and created newspaper
9:23 am
online. his latest effect is something called smart news delivered by app, it has 15 million readers. here is a tv ad that illustrates that point. >> smart news, let's make america great again. >> yes, we can. download smart news. >> see, i was right. >> first time for everything. >> rich, our chief journalist there where we works and tweaks the algorithm. do you know why those two people were so happy to talk to each other? >> they were actors. >> bubble, we use that word. define to me the bubble that you are trying to expand. what is the bubble? technology has made it possible to deliver scientifically proven content to you that is
9:24 am
incredibly interest to you and only to you. and, that's a bad thing. that's kind of like serving you up crack. you keep on consuming more and more of it and your life deteriorates. >> if you get angry in you have of somebody on twitter feed and you stop following them and ureyo you reenforce your own set of believes. we are trying to do is use the technology to expand your perspectives and compel you to get different points of views even points of views that you may not agree with but at least you can hear what other folks are saying. >> let me ask you a question on that, today how this is been done, i clicked on something and i read it. that reenforces the model and the learning. how is that all work if you are not taking into account my feedback. is that a global universe of
9:25 am
feedback. here is how many things that you kick in that are on this. i need to give you more of that. we do take into account of your feedback because we want to know as much as possible of your interests so we can find things that you didn't know you would appeal to you. it is not just about politics. but, it is about sort of what your level of interest is in things that you have not necessarily have. >> this is why i like traditional media. i would not ever want to listen to a report about it. >> it is kind of interesting. because i have chosen the story myself, i am forced to listen to it. oh, i didn't know that. one of the stories that i came across smart news a couple of months ago that i just loved was an australian story was efforts to preserve tazmania devils.
9:26 am
the only one i know is a cartoon. nothing i ever capped on directly said rich would read this story. yet, i think because what i tapped on and sort of breath of my interest, the algorithm made that kind of educated guess that maybe i would be interested in this. i read every column of that story, it was delightful. >> how are you making off money, the app is free and content is free. >> the other piece of what we do is we want to be good partners with publishers and content providers. we have ads that we sell within smart news but we have ads and ways to drive revenues to our publisher partners. we are trying much to nurture the ecosystem. if the ecosystem does not
9:27 am
succeed. we'll not succeed. >> it is a free app, smart news. you can download it. thank you very much. back in a moment.
9:28 am
i don't see the commercial breaks from here in the studio but i am going to presume we show you an app for our new podcast at least one time. not now to blame the point. we got a new podcast and i want you to subscribe to it. you can download it to every device you have including the old iphone you have in your junk yard. download it and you will get your podcast. thank you to my guests and thank you for making us part of your sunday morning.
9:29 am
9:30 am
hello to community dad. nbc bay area presents comuni-dad. with damian trujilo. a very fascinating segment that we should be doing a lot on the show. >> thank you. i am happy to be here again. >> thank you, tell us who we are going to feature today? >> today we are talking to san francisco to meet jorge


info Stream Only

Uploaded by TV Archive on