Slashdot Mirror
Thai Students Score a Prize For Speech Software
Julie188 writes "A team of four Thai students beat out 10,000 competitors to win the $25,000 prize in the Microsoft 2007 Imagine Cup. Their project is text-to-speech software in which computers read aloud typed and handwritten commands. The software will allow people who can't read to interact with a PC. Imagine Cup judge Rand Morimoto has been blogging on the whole experience — from his video of the opening ceremonies to how contestants swilled free Cokes to keep themselves awake during the 24-hour, no-sleep phase of the competition."
Like this?
-:sigma.SB
WARN
THERE IS ANOTHER SYSTEM
"The software will allow people who can't read to crash Vista..."
I am sure that there are many other solipsists out there.
I'm not sure Microsoft should be gaging text->voice software with their track record with voice->text software. http://www.youtube.com/watch?v=2Y_Jp6PxsSQ
"Always forgive your enemies; nothing annoys them so much." - Oscar Wilde
I can't find anywhere in the article mentioning about the Thai students.....
His exploit "just works". Apple fanbois everywhere implode in a self-collapsing vortex of cognitive dissonance. by jjack
I wonder what this system says when the input text is "Dear aunt, let's set so double the killer delete select all"?
Slightly disreputable, albeit gregarious
let's settings sohh daobul thah killa deleting selected auull
Cancel or Allow?
"Always forgive your enemies; nothing annoys them so much." - Oscar Wilde
Rob Miles a lecturer from Hull, UK has also been
5 0/
blogging the event.
http://www.robmiles.com/
http://www.flickr.com/photos/robertmiles/11037035
And in other news, Microsoft sponsors 5,000 Thai programmers with H1B visas. Microsoft also announced today the 'temporary layoff' of 7,500 current programmers. Company accountants claim this move will save the Company approximately 25 million dollars per quarter, allowing it to further aquire intellectual properties ranging from 'the wheel' to 'the zipper' to 'velcro', products that should increase the Company's bottom line to ludacrous profit margins...
Understanding the scope of the problem is the first step on the path to true panic.
Imagine Cup home page
Press release about the winners
What I'm listening to now on Pandora...
Microsoft announces that due to the success of the "let programmers drink loads of coke to induce 24h no-sleep phases", this has now been implemented into the standard work-week at Microsoft to increase production.
They now hope to have Vista SP1 out within the next 48 hours, and while SP1 is installing it will now speak out what it patches.
The software will allow people who can't read to interact with a PC
First you bring VB to the world and let those who shouldn't develop ANYTHING software wise do so... now your plan is to let idiots who can't even read to use a computer? And we wonder why the computing world is a bog of what it once was...
"When life gives you lemons, don't make lemonade. Make life take the lemons back!" -- Cave Johnson
Thai students......speech software....there's a joke in there somewhere
for every complex problem , there is a solution that is simple , neat , and wrong.
At google, there's no unhealthy food around at all. All the drinks are smoothies, fruit drinks, etc. And all free.
Interesting difference in culture.
(Actually I think there might have been a few cans of fizzy drink in the cafeteria. Can't quite remember)
correct me if I'm wrong but isn't that what the narrator in WinXP does? How is this new and innovative? Text-to-speech has been out for years!
Windows Accessibility ++
The game.
Source code please? If it is not available then they are a bunch of lametards.
The same reason you reward a student with a C average who gets a B, and try to encourage a A average student who gets a B to do better next time.
lets see get 10,000 developers to write us a piece of voice recognition, make it a prize and pay them 25k then reengineer it and sell it for HOW MUCH?
Feel the Luv, as Microsoft challenge the open source community on patents.
(OMG, I said Microsoft was not GOD of Software that's a Karma hit)
For better free-as-in-beer text-to-speech, try scribd.com. If you upload some text there, they'll automatically make an audio version, and I thought the quality was amazingly good. (If the text is copyrighted, you can set it to be available only to yourself.)
Find free books.
What's interesting is when Google does it, It's OK. When Microsoft does it, it's evil. Nice to know I'm hanging with the winning crowd. Now if you'll excuse me, I have a hate group to join, so I don't feel all dirty.
It say me rikee the flied lice so much, it make me so hohney, me ruv you rong time
They could have bought a 1992 Macintosh.
I hope they offer Vivarin smoothies.
https://www.eff.org/https-everywhere
"how contestants swilled free Cokes to keep themselves awake during the 24-hour, no-sleep phase of the competition"
why is such a phase needed? i could see it in an extreme gaming tourney, but coding? free coke for your software ideas, from a multi billion $ company?
i guess there is a reason they are rich. they dont even bother to steal finished products anymore, they get young coders hooked on a couple of free snorts of coke
and work them to death. heck this type of labor relations built the pyramids, the cotton industry, and look how much $$ it saved the nazis.
am i missing something here? or is anyone else horrified by this concept?
25 grand is a lot more relative money in Thailand. That's like a year's salary if not more there. This is kind of outsourcing in disquise.
Table-ized A.I.
With about 30% analphabetism and double that for functional illiteracy - this is great news for all those US-Americans who used to rely on television.
This enables people to learn to read without additional teaching resources. There doesn't only have to be a single solution to a problem. Nobody learns to read overnight, this would help those still learning to read during their years of study. With handwriting recognition, it could help people learning to write as well. It has potential to be the tutor for when your teachers aren't around.
The thing is, TTS which is great for visually impaired users has been around for decades, and all these fancy new systems are no better, in fact they're worse.
Listen to something like AT&T Natural Voices which is diphone based, and really no good for VI users as you can't use them at any great speed and understand them well.
Compare that with some hardware synth from the late 80's or 90's, or a software synth like eloquence and hopefully you'll see why the not-so-human-like voices are much better for the people who really need them.
Of course for automated phone systems and GPS navigation, the human-like voices are good, but you need a lot less information from them, try listening to a book, or the contents of your browser window. A lot of commercial screen readers come with Eloquence, and those that don't usually come with something similar, and for a good reason.
While I'm on this point, I wish that somebody would develop a good TTS engine open source, festival is good for what it is, but it's built like the AT&T or Cepstral voices rather then a purely synthetic synth. Ah well.
They should read the terms of the competitions MS runs. It gets to own all the rights to anything submitted.
...computers read aloud typed and handwritten commands. The software will allow people who can't read to interact with a PC They'll still need to be able to write, though. Of course this has its uses for the visually impaired.
Visit http://ringbreak.dnd.utwente.nl/~mrjb/growingbettersoftware to download your free copy of the book
I'm sorry that you were modded down as a troll, but you do have a valid point.
I was at the Imagine Cup competition in Japan some two years ago, and one of the things that was pissing everyone off was that besides the meal times (breakfast, lunch and dinner) you couldn't get non-sweet food. Not even plain bread was available, even if you asked the support staff from the hotel.
On the other hand, chocolate-filled cookies, sweets and all kinds of energetic drinks were freely available in quantities. They did have water, though, so it wasn't completely horrible.
So yeah, I had exactly the same thought during the competition: Someone at Microsoft wasn't really concerned about the student's health, or was thinking something along the lines of sugar = energy = productivity.
BTW, not to bash entirely on MS. Other than on the snacks aspect, they treated everyone really well, and it was one of the most amazing experiences in my life.
A slashdotter who didn't build his own computer is like a Jedi who didn't build his own lightsaber.
Mine got to your sig and... well... I guess it makes sense that shellcode would look like perl.