above the earth. scientists wants to see how the human body copies copes with longer stays in space. any time on our website. the address for that is aljazeera.com and watch us live by clicking on the "watch now" icon. aljazeera.com. >> the frustration of a synthetic voice for those who can only speak through computers? >> it sounds like a robot. >> researchers creating sounds that capture individual pers personali personalities. giving a face. >> what about your speech. >> my tongue. >> for those in need, new voice technologies that over the sweet sounds of success.

>> it is funky. >> yeah. like you. >> this is "techknow" a show about innovations that can change lives. >> genetically modified food. >> we are going to restore the intersection of hardware and humanity, and we are doing it a new way. this is a show about science by scientists. "techknow" investigators giving voice to the voiceless. >> welcome to kechknow. i am phil torres. we meet someone knew, they will often make immediate decisions about who you are based upon your general appearance and what your face looks like and what you are wearing but, also, on the sounds of a your voice? >> it's true.

there are tennessee of millions who can't speak or have limited speech. two and a half million of them are right here in the sglurns. >> of those two and a half million, 40% use some kind of text to speech to communicate but they all sound pretty much alike. >> at the young's home, this holiday season was a turning point. that's because 16-year-old sara young was about to get a gift like no other. her very own voice. >> she has been known as savvy sarah, happy and joyful. >> sarah was born with cerebral palsy leaving her unable speak but had he she and her mom have plenty to say. >> hello, everyone. my name is sarah.

i am 16. >> sarah talks mostly through a text to speech app on her computer . >> i am see it. >> the computerized voice is heather. it's standard equipment. >> why was the heather voice inappropriate for sarah? >> it had an older tone to it, speaking in this if thely-formed female adult voice with a digital sound to it. it was funny, weird hearing it to an unfamiliar listener, it kind of became white voice. >> the voice isn't sarah's alone. you hear it and think she is not here. it was somebody else, you know, from school or someone else with that voice. >> my name is meg. i am 9 years old. >> if you think that that will sounds a lot like sarah, you are right. it's because they both use the heather voice.

>> my voice isn't me. it sounds like a robot. >> may's parents completely agree. >> she doesn't have a voice that sounds like her. you see her eyes and you see her smile, the way she moves and looks at you. you get so much from her. it's. >> it's not her own personality it's still an adult voice. >> do you want your own voice, may? >> yes. >> what do you think having your own voice will do for you? >> it would be cool. i will sound like a kid. >> creating a viable way to build unique voices for those unable to speak is the brain child of speech scientist ruppel patel, founder and ceo of vocal id. >> we are going to go back in that booth and have you say some vowels like you did last time like ahh. can you say that now? >> ahh. >> beautiful what we are doing

at vocal id is we are taking a tiny sample of someone's voice and hours of someone else's voice. >> let's do it another time. let's do it another time. >> we are trying to make it sound like the person who the tiny sound was from. that's the scientific challenge. >> with thousands of english speaking donors from around the world, ruppel and jeff, the director of research and technology, are creating the voices. >> can you kind of walk us through how a voice is made? >> yes. this is sarah here this is an example of what she gave us. >> hum. >> sarah's natural voice? >> that's her voice. when we speak, there is air in our lungs. we force it out. it goes through our trachea and hits the larynx and the vocal cords and those vibrate and from that produces a sound.

it's shaped by your tongue and knows. she can't properly shape her vocal tract to produce each. we separate the vocal tract and we take the donor and do the opposite. we toss away what's going on in the vocal cords and keep going on with what their tongue and lips are doing. >> like an online dating service, the computer matches acoustical patterns with the donor and the vocally challenged recipient. >> we train the synthesizer. >> what do you mean by that? training it? >> we are training it on different parts of speech. up here, it's recording which parts of speech have been captured and like the basic united of speech. in english, there are about 40,

i think. this here, ai and this here is an th south. >> foiptd give a voice? >> this is our home page. from there, click on it. give a vis. you can say, "ahh". >> it's ready to record. record and speak. >> every now and then, you surprise me. >> stop and play. >> every now and then, you surprise me. >> a full donation runs 3500 sentences, about four hours. to capture all of the possible sounds needed for speech. >> let's do it another time. >> let's do it another time. >> 2000 miles from sarah down in austin, texas, 16-year-old lia white was sporadically donating her voice. then she got an e-mail from vocal id. >> they said we found a match for your voice.

i actually printed out sarah's pictures right here, and i hung it up on my bulletin board to remind me donating. this was what they sent me as an e-mail. there was someone on the other end that motivated me to get that done. lo and behold, sarah's voice is now with her. yay. >> when "techknow" returns. >> hello, how are you? >> after years of hearing heather, sarah is going to get a voice of her very own. >> how do you feel about getting rid of heather? >> bye-bye. i will bring the old away and now the new. >> we want to hear what you think about these stories. join the conversation by following us on twitter at aljazeera.com/techknow. >> they don't want anyone to know what is going on inside.

>> he was not just a person, he was a human being. >> "faultlines". >> what do we want? >> al jazeera america's hard-hitting... >> today the will be arrested. >> ground-breaking... >> they're firing canisters of gas at us. >> emmy award-winning, investigative series.

this is a big day for sarah young, a big day, too, for the team at vocal id. speech scientist rupl patel and jeff meltzner are giving sarah a voice of their own that they created. >> how do you feel about heather? >> bringing the old out with the

new. >> the "star-telegram" standardized choice from her text to speech computer program used by tens of thousands of talkers. >> how did you come up with the idea for building voices. >> in 2002 i was at a conference in denmark and i saw a girl and a man using different devices but the exact voice. i saw that happening all around me. >> what's groundbreaking about sarah's voice is that itsnique milking the donor speech with the few sounds sarah can make to capture sarah's personality. >> is there any art industry or psychology involved in creating a voice? >> yeah. i think every science has an art form to it as well. in the case of making a custom voice, it's not only about is it understandable? it's about is it authentic and does it resonate with the person using it? do they feel like it's them? so sarah, amy, what we have done is gotten two voices ready for

you today. and we want to play them and whichever one you choose, we will load to your device. >> voice 1. >> hi. my name is sarah. i am 16 years old and i.. >> what do you think of that voice? do you like it? >> a little bit. >> here is number 2. >> hi, my name is sarah. i am 16 years old. >> 2. >> goodness. >> what do you like about 2? >> it is funky. >> yeah. like you. >> do you want it loaded on your device or think about it, heather? okay. that's the next thing we can do. all right. okay. good how do you feel?

>> awesome. >> there is a feeling you feel. >> the word swat spunky to describe the voice. she feels spunky. >> she is spunky and the world needs to hear her. >> your machine. >> i am here. you can do it. >> i am glad you knew that. >> how does it make you feel? >> awesome. >> what do you feel hearing your voice? >> surprise. >> hey, sarah. >> i was watching you, you chose your voice. i also like the second voice. >> a happy girl. >> now, you have a sassy voice to go with it. >> the real test would come the

next day at the school in lexington, massachusetts. >> good morning, sarah. are you excited to show your classmates your new voice? as we headed to sarah's class, i was nervous for her because we all know how critical high school kids can be. >> sara is getting a new voice. right? her voice is individualized. no one else has it, like our voices. right? no one else has our voices. so you have the same voice as sarah. right? could we hear your voice? >> hi. my name is elizabeth. >> can someone describe what that sounds like? >> it sounds pretty good. right? >> like siri on your iphone. >> are we ready for sarah's new voice? >> hello. my name is sassy sarah.

>> how are you liking it? >> awesome. >> can we hear something else? >> i like things that a lot of girls like, like sleepovers, chock look at, high backseats. >> with a show of hands, who likes this sarah's new voice better? all right. >> having your own voice shows your personality. >> sarah is not the only one at school with a newer voice. only nines years old, maeve was using the new voice. >> my voice sounds like a robot. >> maeve's sister made the voice. >> my coins, the first thing i want to here maeve say is "i love you, aaron". >> and maeve's dad new what he

wanted to hear? >> "i love you, daddy." i would love to hear that in a voice i recognize as maeve's. >> hi, maeve. i am lindsey. maeve, you have had your new voice for about a week now, i think. tell me what you think of it. >>ty like my new special voice. i can talk to everyone. everyone is excited to hear my voice today. >> i heard your dad say that he was really looking forward to hearing you tell him that you loved him using your new voice. so have you done that yet? >> kind of. >> kind of? why is having a unique voice so important? >> it makes us human. we make judgments about people.

i feel like it's very important for people who don't need devices to talk to have that same ability to project who they are and to be known as people and not just robots that talk. is. >> i hope i helped you out. >> that was an incredible piece. it shows you the value of donating what you have. you don't think about voice donations? >> i don't think we think about how important a voice is. one of the most moving moments for me was watching sarah debut her voice for her classmates and having them discuss how important it was for each of them to have a unique voice. particularly for those with limited speech. they all cited having their own voice as a way not tonal communicate but to be able to fend for yourself, speak for yourself. it's really such an important

part our personality did we don't think about. >> your voice is your number 1 advocate. so especially within this disability community, how feasible is it? is it acceptable? is it expensive? >> right now, it's expensive, about $1,250 to develop a voice through vocal id but the hope of the company is to make it cheaper to make it more accessible and potentially to be covered by health insurance. >> i think what is interesting is yours is about improving upon the past conditions. you have a storyb preparing for the future. >> how to take care of your voice before it's long gone.

we have met one of millions of people who have a limited ability to speak. technology is there to help them. kara in your case, you are about to tell us about somebody who knows in the future they will lose their voice. >> a touch position to be in. al.l.s maybe throat cancer. there are a lot of degenerative

conditions in which you know you may lose your voice. a new program, my own voice, helps you capture the essence of your voice so yournique characteristics shine through. let's check it out. >> i was diagnosed officially in july of 2014 with a.l.s., also known as "lou gehrig's disease". >> she is a reporter for rice media. >> in simpler terms, you completely lose all function, but you feel everything, and you are aware of everything. so, it's like getting trapped in the quicksand and losing a little bit of yourself every day. >> what kind of symptoms ri dealing with? >> now, i lost most of my fine motor skills. it's almost impossible to type. it's frustrating. >> perhaps the most cruel loss,

the ability to speak. >> what about your speech? >> i do notice that my tongue is starting to twitch which is a sign that the muscle is looking for that signal from the brain and not getting it. >> how would you describe your life? >> when i came to the u.s., i learned english watching t.v. >> specifically, the simpsons with closed captioning. my speech is probably just run-of-the-mill, midwest, what you hear on t.v. >> is he the only guy you have left in are you trying to get through the day with less sleep? >> to prepare for the inevitable silence, she is recording the 1500 sentences for a revolutionary text to speech program called "my own voice." it creates a synthetic voice that will capture the essence of her speech. >> why did you think that that would be a good idea? >> i usually, at first, i didn't think it was a good idea. >> could you write another script for us?

>> your voice and how you communicate in the and the cadennce which you say something, i am going to completely lose that. that's why i can't be excited about a robot voice. >> his reaction to the article? >> yay! now, i am actually done. that's better. >> maggie is an assistant technology speech specialist who is evaluating "my own voice". >> a choice of shoes to wear. he is big, the way water buffalo are big. >> the sentences literally make no sense. >> no. >> because with these phrases, they are getting all of those sounds that they need. >> it fills the voice on the gramat cal rules and the pronunciation. so you will never hear these sentences good. the software is going to take all of the sounds and chunk them all out. >> vincent paoel explains how it

is done. >> impress sglif it actually comes from the word . >> impossible comes from the word? >> president. >> and if i take the last syllab syllable, it comes from...? >> detective. >> they are going to make the word impressive. >> an ocean away from acapella, angelina is ready to hear her stand-in. my. >> okay. so the sentence i typed in was, i came here expecting cup cakes. i have very little expectations for this. >> i came here expecting c cupcak cupcakes. >> that's kind of creepy.

it sounds just like me. the kadance is a little bit off. the actual tone is exactly like me. angelina tried a sentence? >> this will say, hello, i am angelina and this is my robot voice. >> hello. i am angelina and this is my robot voice? >> sounds a little awkward. it sounds like i turned into a robot. i wanted to hear robot me, although i only recorded about half of the required sentences. and as a refugee from college -- i screwed that up. the following week when i logged in to get my voice, my expectations were low. my dog's name is killer. i am going to call him. >> killer, here, boy. come to your mama . >> i am not impressed. i don't think he is coming.

>> he doesn't come when i call him either. >> a lot of people don't realize actually the first synthetic voice was made for stephen ha hawking for ibm. in some ways we identify with him but not aefshl would wants to sound like him. >> or like heather, which is the synthetic voice that most fem users have. >> it will be interesting to see. even in the cases of sarah and maeve. they are appropriate to a 9-year-old girl and a 16-year-old girl. what's going to happen when they age out of those voices? >> you know, this idea of voice donation, of blending the sounds you can make with the donated

voice or voice banking, is a technology that's wild. it's amazing but i don't know if the technology is quite there yet. >> there is a lot that goes in to a voice, the sides of your lungs, the shape of your throat, the muscles in your tongue and your face. all of these things are quite complex so at this understandable that a software, a synthetic thing that is not a human, it human, it's it struggles to sound exactly like us and to sound natural. >> maybe 1500 sentences isn't enough. maybe what we could be doing in the future is taking every recording we have ever had and blending it in to a voice to use for posterity. that's a cool thing about living with modern technology. there are recordings of everybody because we use our phones. >> here on the show, we have met people like in this case who are trying to make life with these conditions better. we have also met scientists and innovators trying to cure these conditions. it's two battle fronts. interesting stuff. thank you for your stories.

that's it for "techknow". >> go blind the scenes at aljazeera.com/techknow. follow expert contributeers on twitter, instagram, google+ and more. this is al jazeera america. here are today's top stories. protesters in arizona blocking a road leading to a donald trump rally but it did not stop the republican frontrunner from speaking to his supporters. cuba is preparing for the rival of president obama tomorrow. the first time a u.s. president visited the island nation since 1928. a sci