Beta

×

Welcome to the Slashdot Beta site -- learn more here. Use the link in the footer or click here to return to the Classic version of Slashdot.

Thank you!

Before you choose to head back to the Classic look of the site, we'd appreciate it if you share your thoughts on the Beta; your feedback is what drives our ongoing development.

Beta is different and we value you taking the time to try it out. Please take a look at the changes we've made in Beta and  learn more about it. Thanks for reading, and for making the site better!

PDA Speech Translator

michael posted more than 10 years ago | from the someday-it'll-cost-$9.99 dept.

Handhelds 161

jlowery writes "Not quite as good as a babelfish, but a PDA that does translation is probably better than resorting to hand gestures alone. I could see this as a boon to the tourist who travels to places where English speakers are uncommon."

cancel ×

161 comments

Sorry! There are no comments related to the filter you selected.

Don't forget... (-1)

SCO$699FeeTroll (695565) | more than 10 years ago | (#7839793)

...to pay your $699 licensing fee you cock-smoking teabaggers.

I forgot (-1)

Fecal Troll Matter (445929) | more than 10 years ago | (#7839806)

to backup my pr0n. now i'm out. what do i do?

-hard in houston

Re:I forgot (-1)

CmdrTaco (troll) (578383) | more than 10 years ago | (#7839822)


Well, if you're gay, stick around. I'm sure much faggotry will ensue.

Re:Don't forget... (0)

Anonymous Coward | more than 10 years ago | (#7839835)

You know, I really admire your consistancy.

fp suxxxxxxxassssssss (-1, Troll)

Anonymous Coward | more than 10 years ago | (#7839810)

LOL CATO! LOL!

dear sir (-1, Offtopic)

Anonymous Coward | more than 10 years ago | (#7839864)

I could simply point out that YOU FAIL IT and move on with my day, but your FAILURE strikes me as a particularly pathetic one. Not only did YOU FAIL IT, but YOU FAILED IT with absolutely no style. There's nothing funny or offensive about your FAILED attempt. You came, you proved that you are teh sux, and that was that. Consider yourself lucky that you're getting a reply at all.

Next time YOU FAIL IT, try to do it with some panache.

Re:dear sir (-1, Offtopic)

Anonymous Coward | more than 10 years ago | (#7840001)

You, sir, have offended me by omission.

Earlier today I failed it (narrowly) with the mention of Lore Sjoberg.

You FAILED to mock me as you mocked the grandparent.

I am deeply hurt by your insensitivity.

FP (-1, Offtopic)

Anonymous Coward | more than 10 years ago | (#7839824)

FP

Re:FP (0)

Anonymous Coward | more than 10 years ago | (#7839947)

You suck so much at getting FP that you can't even get First FAILED IT!!!

The problem with these things (3, Insightful)

the man with the pla (710711) | more than 10 years ago | (#7839830)

The problem with every software that I have used that tries to decipher human language (like Zork or the game included with emacs for X) is that you have to know what words the software understands and in what context.

I have seen the same problems with automated phone systems that are supposed to recognize a generic voice and I can see the same thing happening here.

The main difference here though, is that when entering text, you know exactly what you input before pressing enter. With voice recognition software, how do you know that the software "hears" exactly what you say? If you say somethign like "What are my appointments for the thirteenth?" and it hears, "What are my appointments for the thirtieth?" you would be receiving the wrong information.

I hope this is a success but I don't have my hopes up.

--
7329756

Re:The problem with these things (2, Insightful)

garcia (6573) | more than 10 years ago | (#7839887)

My father has a 2004 Acura TL with Bluetooth cellphone stuff... He was trying to get it to dial a number. What a pain in the ass. It was seriously almost as distracting as hand entering the number. I believe he had to ask it to dial XXX-XXXX 5 or 6 times before it stopped adding in two random zeros.

Until the machines can be 100% accurate without frustration they are next to useless.

Re:The problem with these things (1)

ackthpt (218170) | more than 10 years ago | (#7839913)

The prototype falls short of Star Trek's fictional universal translator in several ways. What? How? ... no way!

How do you enter umlauts and other characters in Graffiti?

Hola! I would like some fried fathers!

Re:The problem with these things (1)

jpmkm (160526) | more than 10 years ago | (#7839998)

For an umlaut, first do a vowel, then do a w with a horizontal line attached(w-), and don't pick up the stylus between the w and the line.

Re:The problem with these things (1)

shuz (706678) | more than 10 years ago | (#7840046)

There are 3 main reason why this happens. Processing power and enunciation and possibly the quality of the microphone. The algorithm used to determin what numbers are being said is most likely based on common pronunciations of numbers. Dial is a pretty easy word to figure out because it starts with a 'da' sound and ends in a 'll' or 'el' sound. What happens in the middle doesn't matter too much. some numbers may be harder to figure out such as five which could be pronounced as fif fiva fiev ect. Also once a word has been recorded the algorithm uses statistical analysis to further figure out what was most likely said. So that if the word fiff was recorded the computer knows that its most likely not six because the Sss sound is not very close the recorded word. The processor has to do this statistical analysis in realtime which means the algorithm can only be run a few times on each word and a hit of only 70-80% may be ok. combine that by 7 to 10 words and you have yourself possible poor performance. Of course the microphone could have a lot of noise in it which just futher throws off calculations.

Re:The problem with these things (3, Insightful)

Angus Prune (660032) | more than 10 years ago | (#7840047)

It all boils down to confidence. I have to be confident that what I'm doing will work.
I use a wireless keyboard but Im having to switch back because I find I have to check what I am typing because it doesn't always pick up every keypress
Voice to text are only of limited use while you have to re-read and correct any mistakes.
While this is only 80% accurate it can never be trusted. When this works at 95% it won't be trusted. I won't trust that this won't mistake Renal for Venal.

While this is a great step foward I can't see it being trustworthy for 2006 and I still think the same problems still apply to this as have always applied.

Re:The problem with these things (3, Insightful)

fastidious edward (728351) | more than 10 years ago | (#7840265)

It was seriously almost as distracting as hand entering the number.

Are you being sarcastic? I can type a number on a numeric keypad much faster than I can say it. The 5-6 times much more than compensates for the time of getting the phone out of my pocket.

Voice recognition is great, but tactile recognition is also great, as is body movement.

Until the machines can be 100% accurate without frustration they are next to useless.

I know I have trouble understanding someone with a heavy Southern-USA accent, like someone else may have trouble with a heavy Scottish accent (as firends have) or heavy London accent (as I can revert to), people are not perfect at understanding people, let alone machines understanding people.

Voice regognition is not a great saviour and IMHO is years away, in the meantime I'm happy with a numeric keypad.

Re:The problem with these things (1, Interesting)

Anonymous Coward | more than 10 years ago | (#7839890)

***I have seen the same problems with automated phone systems that are supposed to recognize a generic voice and I can see the same thing happening here.***

Yeah I see it all the time, we who have a SLIGHT southern twang in out accents give those voice command systems fits.

Re:The problem with these things (1)

fastidious edward (728351) | more than 10 years ago | (#7840056)

I believe the solution may be:

"...device for translation of spoken languages in real-time communication. It operates by scanning brain-wave frequencies and using the results to create a basis for translation". A la the portable universal translator." More here [aol.com] .

Personally I'd prefer my own Ensign Hoshi [startrek.com] .

Re:The problem with these things (0)

Anonymous Coward | more than 10 years ago | (#7840412)

Yeah but compared to the miscommunication you'd get otherwise with hand gestures and mis-understood words? I've done some travelling, I know a little French and it can be really frustrating and worse in places where you only know a few basic words.

If it gets cheaper, I think it could be a big hit amongst the backpacker crowd or at international events. You could probably learn the language a lot faster too, which could potentially increase the overall number of languages people can speak or communicate in, which I think would be cool.

huh? (1, Redundant)

xie (722634) | more than 10 years ago | (#7839838)

The prototype falls short of Star Trek's fictional universal translator in several ways.
What? How? ... no way!

Re:huh? (1)

ackthpt (218170) | more than 10 years ago | (#7839873)

The prototype falls short of Star Trek's fictional universal translator in several ways.

What? How? ... no way!

Uhura: We get signal!

Kirk: What!?

I see what you mean.

Re:huh? (0)

Anonymous Coward | more than 10 years ago | (#7839972)

for one, i'd rather been seen with that hoshi chick on my arm than a pda in my hand! she's awfully purty!

Had to be said (5, Funny)

Anonymous Coward | more than 10 years ago | (#7839839)

"All your base are belong to us!"

Re:Had to be said (1, Offtopic)

valkraider (611225) | more than 10 years ago | (#7839926)

Moderators, of which I have been one from time to time, need to do their research before modding. Or at least pay attention...

I was about to say "All your Base are belong to us!" when I saw this post. You see, moderators, the "All your base are belong to us!" comes from a poor translation of a video game - and poor translation is not OFF TOPIC. We are talking about a translating device - which would give us potential real world "All your base are belong to us!".

Mod parent up +1 funny, and +1 insightful (Too bad it is an AC posting)

MOD PARENT DOWN (0)

Anonymous Coward | more than 10 years ago | (#7840081)

He's being demeaning to you moderators and questioning your judgement.

Good Grief... (3, Insightful)

ackthpt (218170) | more than 10 years ago | (#7839842)

I could see this as a boon to the tourist who travels to places where English speakers are uncommon."

Spoken like someone who has never taken a foreign language class. Suppose that thing is going to get the accent right? Emphasis on the right syllable? Not likely, mostly good for translating some text message into the PDA holder's tongue (and doing an Engrish [engrish.com] job of it anyway.)

Re:Good Grief... (2, Funny)

UrgleHoth (50415) | more than 10 years ago | (#7839876)

Reminds me of the joke:

What do you call someone who speaks three languages? A polyglot.
What do you call someone who speaks two languages? A bilingual.
What do you call someone who speaks one language? An American.

Re:Good Grief... (0)

Anonymous Coward | more than 10 years ago | (#7840028)

What do you call someone who doesn't speak English?

Atavistic.

Re:Good Grief... (2, Insightful)

NanoGator (522640) | more than 10 years ago | (#7840090)

"What do you call someone who speaks one language? An American."

I know it's a joke, but it's a common complaint aboout Americans. Unfortunately, nobody seems to think about the United States' geography and why most of us are uni-lingual. To the North, we have Canada, which is mostly english speaking. To the south, we have Mexico, which is Spanish speaking, but there's not all that much travelling back and forth like there is with Canada. Worse, they're very accomodating down there, so there isn't a big huge need to speak Spanish. Go much further south than that, and you're spending a great deal of money to get on a flight to do this. (I should know, I've traveled to Brazil twice.)

This is very different from Europe where you can drive across countries like we can drive across states here. Even if we were bilingual, there wouldn't be a huge screaming need to speak in other languages. It's hard to feel the need to speak other languages when you have to travel overseas to encounter somebody speaking that language.

Sadly, this factor is never considered. Nope, it's assumed we're just stupid.

Re:Good Grief... (0)

Anonymous Coward | more than 10 years ago | (#7840172)

I suppose it boils down to, whatever country you're in:

1. Are you happy getting by?
2. Are you interested in the challencge a language can bring?

As you say 1 can lead to learning a second language. This can lead to 2. But we may never know.

IMHO Americans not learning Spanish is damn insular and imperialistic, they are your neighbour, not your slave, so why not put in some effort and try rather than assuming they are accomodating?

Re:Good Grief... (2, Insightful)

ackthpt (218170) | more than 10 years ago | (#7840232)

I suppose it boils down to, whatever country you're in:

1. Are you happy getting by?
2. Are you interested in the challencge a language can bring?

As you say 1 can lead to learning a second language. This can lead to 2. But we may never know.

IMHO Americans not learning Spanish is damn insular and imperialistic, they are your neighbour, not your slave, so why not put in some effort and try rather than assuming they are accomodating?

Allow me to be cynical here. People cowtow to the language of commerce. If a lot of german people with a lot of money are visiting your town, you can bet people are learning german to be accomodating. This has much to do with why japanese, chinese, germans, belgians, dutch, italians, indians (asian) and even french, learn the language. The question is, with the blossoming of China's economy, will people turn to learn the business language of China?

Re:Good Grief... (1)

NanoGator (522640) | more than 10 years ago | (#7840359)

"IMHO Americans not learning Spanish is damn insular and imperialistic, they are your neighbour, not your slave, so why not put in some effort and try rather than assuming they are accomodating?"

It has nothing to do with being insular or imperialistic. To assume so is a bit ignorant. It has to do with how useful it is. I do, more or less, speak Spanish. I so rarely need it. Seriously, the most mileage I've gotten out of knowing any Spanish (outside of a trip to Brazil, where it was only a minor help as they speak Portuguese there) was the understanding a skit I saw on Mad TV.

Let's put this another way: Could you go an entire year without wearing shoes? You could, but you'd never dream of it. Why bother? You always have shoes, they always protect you. You have no dying need to have your feet harden from walking over gravel. How would you feel if I grouped you in with people who couldn't live without their precious shoes? You'd feel like I was a moron for making that distinction, and an even bigger one for not considering that you haven't run into a situation where you never have shoes.

Think about it.

Re:Good Grief... (0)

Anonymous Coward | more than 10 years ago | (#7840213)

Well, what do you expect? English is the most common of only three languages in common use on the entire continent. In many areas of the US, you can travel a thousand miles in any direction without needing a second language. It's not like Europe, where a continent a fraction of the size of our country is host to dozens of languages.

As for the non-English areas in North America, most (bordering on "all") Quebecis speak English, and if their government dropped its rules on the primacy of French, it would be as Anglicized as the rest of Canada within a decade; and you can get by pretty well with English only in most of the tourist areas of Mexico.

(As a side note, this also applies to Michael Moore's idiotic criticism about Americans not having passports; the fact is, I can travel thousands of miles without even needing to show ID, much less an actual passport.)

Re:Good Grief... (1)

gl4ss (559668) | more than 10 years ago | (#7840030)

yeh and I'd suppose that you alrady have been able to do that(rough text translation) for couple of years through online connectivity(no need for huge local dictionary)..

getting the right word for the context the word is in can be a real art in some languages as well(and no fucking way you'd be getting a spoken to spoken translation anytime soon on a full blown pc much less on a pda..)!

Re:Good Grief... (1)

ackthpt (218170) | more than 10 years ago | (#7840182)

yeh and I'd suppose that you alrady have been able to do that(rough text translation) for couple of years through online connectivity(no need for huge local dictionary)..

For the languages I already know I prefer to read without translation (German, Spanish, French) as it's good practice. The gaffes in translating are the limitation of the software, not the hardware.

getting the right word for the context the word is in can be a real art in some languages as well(and no fucking way you'd be getting a spoken to spoken translation anytime soon on a full blown pc much less on a pda..)!

The way the capacities of portable electronics are going I see no reason they couldn't, including local dialects and slang. Just get some people who know what they are doing to put it together. Most of what I see looks like textbook translations, i.e. what you get if you pick up a dictionary and do the words one at a time, though they do understand noun-adjective ordering for some languages. Knowing 'papas' from 'paPAs' takes the extra skill and effort.

Re:Good Grief... (0)

Anonymous Coward | more than 10 years ago | (#7840503)

The way the capacities of portable electronics are going I see no reason they couldn't, including local dialects and slang. Just get some people who know what they are doing to put it together.
For someone who claims to know four languages, you sure are way off base about the difficulty of machine translation. The only reason I can come up with is that you must not be in any way familiar with the technical process behind translating and what is and isn't possible now.

When you translate from one language to another, it isn't a matter of matching one word to another, as you probably know. It also isn't a matter of matching one idiom to another. Translation beyond what Babelfish does requires context. And context requires either a form of AI (actually understanding what is written) or a huge amount of sample data from which to statistically determine the most likely path to take in translating each portion of a text (this is, in a sense, simulating an understanding of what is written). This route is what is taken now and it takes a lot of processing power to be able to come up with (somewhat) accurate and fast results. There is a LOT of room for improvement.

A computer has no way of determining if "sa soeur" in a given sentence means "his sister" or "her sister". It doesn't know if "krass" means "gross" or "awesome". It can't pick up things like sarcasm. If a sentence says "Sie ist zu lang", in reference to the word "Hose" in the previous sentence, how does the translation software figure out that "sie ist" should become "they are" when brought into English? And so on. These are just some basic things that pop into my head but it gets a lot more complicated than this.

Re:Good Grief... (1)

Otter (3800) | more than 10 years ago | (#7840212)

Suppose that thing is going to get the accent right? Emphasis on the right syllable?

Admittedly this thing isn't going to get you a job as the Tom Brokaw of Malawi or Laos. But when all you want is for someone to point the way to the beach or the presidential palace, this should certainly be adequate. It's not obvious how it's an improvement over a dictionary or phrasebook, though, except maybe that it gives you sentences.

And here (3, Funny)

Anonymous Coward | more than 10 years ago | (#7839844)

I thought that you only had to speak English slowly and loudly enough for anyone to understand. Silly me!

Re:And here (0)

Anonymous Coward | more than 10 years ago | (#7839968)

As Dave Berry once wrote: '"TAKE ME TO THE AIRPORT", I said, speaking in capital letters so he would understand me.' (or something like that.)

Re:And here (1)

Camel Pilot (78781) | more than 10 years ago | (#7840009)

No no no it is the accent that really matters and gets the point across.

Re:And here (0)

Anonymous Coward | more than 10 years ago | (#7840048)

No, you also have to use a poorly-done imitation of the other person's accent.

Hmmm (2, Insightful)

dreamchaser (49529) | more than 10 years ago | (#7839846)

According to the article, it only works for medical terms so far, and is only 80% accurate. I don't know about the rest of you, but I don't think I'd want to trust any of my medical treatment to such a translation!

Doctor: "Well, we thought he said pennicillin, not omoxycillin! I'm afraid the infection has run amok!"

Re:Hmmm (1)

NeoThermic (732100) | more than 10 years ago | (#7839904)

Or the inverse: We are 80% sure that his PDA thing said that he broke his left hand, however, because of the 20% inaccuracy, we hedged our bets and plasterd his right hand...

NeoThermic

Re:Hmmm (1)

xie (722634) | more than 10 years ago | (#7839936)

I like the thought that since its only 80% accurate they could be spending valuable time making sure the translation is right instead of maybe just diagnosising/treating me.

Re:Hmmm (1)

izzo nizzo (731042) | more than 10 years ago | (#7840061)

for diagnosis this could actually be pretty useful, as it would only have to understand certain phrases such as "where does it hurt" it is disappointing that voice recognition isn't great yet, but be optimistic people. It will be hot when it's ready.

Re:Hmmm (1)

plinius (714075) | more than 10 years ago | (#7840031)

If you had ever dealt with the clueless doctors in some European countries, you would know that every detail you can give them will help. Some of those people, even if the speak English by some chance, seem like they have degrees from community colleges. Medicine is barely more than a hobby for them.

Re:Hmmm (2, Funny)

Spam.B.gone (682244) | more than 10 years ago | (#7840076)

oh no.. he said 'I want a full bottle in front of me'...

Good Idea... (5, Funny)

avgjoe62 (558860) | more than 10 years ago | (#7839850)

> I could see this as a boon to the tourist who travels to places where English speakers are uncommon.

Yeah, I could really use one of these when I go from Fort Lauderdale to Miami...

Re:Good Idea... (0)

Anonymous Coward | more than 10 years ago | (#7840128)

My son is getting ready to choose his High School Foreign Language (after 3 years of intro to Spanish in 6th-8th grades).

He wants to take German. We are in Ohio.

I keep stressing to him how critical Spanish is (we couldn't even order a Gin and Tonic in Miami - that was a tragedy!) Hispanics are the largest minority in the US and certainly the most common non-English language (not counting Kentuckiana language and Java).

What foreign language would you recommend that a High Schooler take?

Re:Good Idea... (1)

avgjoe62 (558860) | more than 10 years ago | (#7840240)

While Spanish is always good to know, in college German or Russian was a requirement for an engineering degree.

Besides, my mom was from Germany, so I had someone to practice with built-in.

If I had a choice, though, and I had the options of your son, I would take German but keep up with my Spanish. Maybe watch Spanish language satellite television channels (aside from the practice, the chicas are hot!)

Yeah, thanks, but I'll wait for a bit... (3, Funny)

dejinshathe (736132) | more than 10 years ago | (#7839853)

"It also works only when the speakers are talking about medical information, and it's only about 80 percent accurate in the lab."

Forgive my immediate misgivings, and you can call me chicken if you want, but I'm really not that keen on walking into a hospital and asking to have a medical procedure done with a 1 in 5 chance that instead of removing my appendix, they might remove my "appendage"...

Re:Yeah, thanks, but I'll wait for a bit... (2, Insightful)

shuz (706678) | more than 10 years ago | (#7840117)

If your willing to not have your speech translated in realtime. Say your willing to wait 5 minutes or so a 95% or better return can be expected. The main reason why these translators aren't accurate a lot of the time is because the algorithm used can only make a limited ammount of passes on each word so that each word is translated in near realtime.

Excellent! (1, Funny)

Anonymous Coward | more than 10 years ago | (#7839865)

Now i can travel to other parts of the USA and be able to understand the locals!

A little bablefish (0)

Anonymous Coward | more than 10 years ago | (#7839874)

English -> French -> German -> English:

Not necessarily also well as babelfish, but a PDA, which makes the translation, is probably better than falling back, in order to give only gestures. She could see this as the favour the tourist, who travels to the places, where the persons of the English speech are a little frequent.

I've always wanted to sound like... (2, Funny)

Anonymous Coward | more than 10 years ago | (#7839879)

...Stephen Hawking in Arabic.

Re:I've always wanted to sound like... (1)

Shut the fuck up! (572058) | more than 10 years ago | (#7839916)

...Stephen Hawking in Arabic.

Just add a little to the name and you got it: Stephen Hawking a luggie.

The military has had these for years (0)

Anonymous Coward | more than 10 years ago | (#7839880)

The prototypes only said "All your base are belong to us" in Iraqi.

audio phrase books (1)

peter303 (12292) | more than 10 years ago | (#7839881)

For when text-books are too cumbersome in the field. I thought these were being used to some degree by the military already.
Like the books they are not intrinsically intelligent.

a complete translator could be possible (4, Interesting)

shuz (706678) | more than 10 years ago | (#7839886)

Technology is at a point where all the software has been written to create a translator where a person speaks into a microphone which then is translated into text which is then translated into a different language which is then played back verbally in the same persons voice in a different language. The problem is that this cannot be done in realtime. 4 years ago I worked on a project for At&t to create an application that would train a users voice, break down thier voice patterns and be able to rearange those patterns to create other sounds which sound like they are coming from that real person. The problem is that with current processors the time to train and process is about 10 hours. So we can do voice recognition in realtime, we can translate text words in realtime, and in 10 hours we can reproduce a persons voice nearly flawlessly. Think of the possiblities!

Re:a complete translator could be possible (1, Interesting)

Anonymous Coward | more than 10 years ago | (#7840138)

So we can do voice recognition in realtime, we can translate text words in realtime, and in 10 hours we can reproduce a persons voice nearly flawlessly. Think of the possiblities!

Yeah, I bet John Ashcroft is creaming his jeans... just think, if your wiretaps don't pick up anything incriminating, you can still use them to *make* someone say something incriminating.

Re:a complete translator could be possible (1)

burns210 (572621) | more than 10 years ago | (#7840534)

so should we look for this in the 3.0 linux kernel? I mean seriously, want to out innovate Microsoft?(i do!) Then lets start work on it. GNU/Speech Recognition, GNU/Translator, GNU/Synthesizer or something like that.

in the 4 years since your work, we have gone through nearly 2 1/2 "Moore Generations" so processors are far beefier than they have ever been. another generation, they will be in the 5+ Ghz range, which should be able to cut that 10 hours of processing down a lot.

Even better, why process it at all? just have RMS GPL his own voice and have in built into the kernel, so it could have common words/sounds already predone, so only a smoother would be needed to make it sound natural... Right?

Could work, in a limited sense.. (2, Informative)

iantri (687643) | more than 10 years ago | (#7839899)

There is a program that already exists for the Palm (unfortunately I do not remember the name) that allows you rudimentary communication with one who speaks a foreign language by translating common phrases, selected by tapping on the screen.

I realize that this software is supposed to be somewhat more powerful, but what I am saying is that even limited translation programs are useful for tourists.

text (2, Informative)

Anonymous Coward | more than 10 years ago | (#7839905)



As speech recognition technology gets better, and as handheld computers get more powerful, audio translators are becoming a more practical proposition.

Researchers from Carnegie Mellon University, Cepstral, LLC, Multimodal Technologies Inc. and Mobile Technologies Inc. have put together a two-way speech-to-speech system that translates medical information from Arabic to English and English to Arabic and runs on an iPaq handheld computer.

The prototype falls short of Star Trek's fictional universal translator in several ways. The system is not transparent -- it must be switched between Arabic-to-English and English-to-Arabic modes. It also works only when the speakers are talking about medical information, and it's only about 80 percent accurate in the lab.

The device shows that it's becoming possible, however, to provide automatic translation using a portable device. "It's good enough to make yourself understood," said Alex Waibel, a professor of computer science at Carnegie Mellon University and a founder of Mobile Technologies Inc.

The effort is one of a series of projects aimed at providing the armed forces with automatic translation for medical and force protection situations and making automatic translation in a wider set of subject areas available for tourists during the 2008 Olympics in Beijing, said Waibel.

The Speechalator prototype uses a built-in microphone and a language-selection button. "You push on the button on the iPaq and speak a sentence and then the translation comes out... in the other language," said Waibel. "You can switch it into the opposite mode when the other person answers and it translates back into your own language."

The software consists of three components: a speech recognizer, a translator, and a speech synthesis engine. "Each one of these components have slight twists to them... in order to work properly for speech translation," said Waibel.

The researchers modified the speech recognition engine to optimize it for handling spontaneous speech.

The translation system has the biggest twist. It extracts the key meaning from the input sentence and translates it to an interlingual, or intermediate representation, and the process depends on the speech being contained in a certain domain, or context, like medical information. "It's just certain nuggets in the phrase that... you need to extract," said Waibel.

The process is akin to constructing a medical-context template that fits the key information, then filling in the template, said Waibel. This process makes it possible for the system to handle spontaneous speech. "We go fishing for the nuggets," he said. But it is also a limitation -- the system must know what domain a speaker is talking about.

The researchers are working on a system that can handle multiple contexts and automatically switch between them, said Waibel. "It can, for example, recognize 'now you're in the hotel reservation domain', or 'now you're in the conference registration mode', or 'now you're talking about medical problem'," he said.

To come up with templates that handle different domains, the researchers collect a lot of data from people talking in those domains, said Waibel. "The more data we collect the better coverage of all the possible ways you could be saying [these things] becomes," he said.

The difficult part was fitting the software required to do two-way translation in the 64 megabytes of memory contained in the handheld computer, said Waibel. "You need two recognizers, two synthesizers and two translators to make [it] happen in both directions," he said.

The prototype also has a camera attachment that translates text like that on street signs, said Waibel. Snap a picture of a sign with the camera and it automatically extracts the text region, puts the text through a character recognition program, then translates it, he said. "What you then see on the screen is the picture of the scene with a sign and then underneath an English subtitle," he said.

The Speechalator is a practical proof of concept, said Bernard Suhm, a senior scientist at BBN Technologies. "They have engineered the recognizers and other algorithms sufficiently to make them work in real-time on the very limited computational resources of a consumer PDA," he said.

The device carries the promise of being useful not only for medical translation, but also situations such as travel or business, said Suhm. "This work could facilitate the transition of speech-to-speech translation research from the technology side of research, which focuses on algorithms and engineering, to the human factors side of research, which focuses on how people interact with devices, and how useful devices are to tasks from real-life," he said.

The device hasn't yet been run through its paces in a field test, however, Suhm said. "Until then we don't know whether the additional challenges in the field, [like] high levels of noise... or usability issues make it unusable," he said.

The researchers' next steps are to increase the accuracy of the device so that it can deal with ambient noise, and expand the coverage by collecting more data about how people communicate in different domains, said Waibel. The researchers are also working on building learning algorithms that automatically sort out different ways to say the same things.

The researchers' next prototype is scheduled to be finished in the summer of 2004, and will initially have two domains: hotel reservations and medical situations. "Then it [it will] gradually expand towards other domains as are necessary for tourists," he said.

The device can eventually be used to provide translation services for soldiers and relief workers in foreign countries and for travelers, said Waibel.

It could also address a medical problem in the U.S., he said. "There are a number of people in the U.S. who don't speak English and then when going to doctors... feel embarrassed to explain their health problems in front of somebody else who translates," he said.

The researchers are also working on a multilingual speech recognizer that can recognize speech in any of a set of languages, said Waibel. "In that case you might not have to switch the system between the two languages -- you just talk in any language and it will come out in any other language you choose," he said.

And they are aiming to develop a system that combines speech translation with human-to-machine translation, said Waibel. "There are certain situations as a traveler... where you want to communicate with a person in another language, but then there are certain other things which you could just as well do communicating with [a computer]," he said. You would want to talk to another person when ordering food, but communicate with a machine to get directions to a railway station, for example.

Longer-term the researchers are looking for ways to deal with spontaneous speech that is not limited to a certain domain, said Waibel.

Waibel's research colleagues were Ahmed Badran, Robert Frederking, Donna Gates, Alon Lavie, Lori Levin, Tanja Schultz and Dorcas Wallace from Carnegie Mellon University, Alan W. Black from Carnegie Mellon University and Cepstral, LLC, Kevin Lenzo from Cepstral, Monika Woszczyna from Multimodal Technologies Inc., and Jurgen Reichart and Jing Zhang from Mobile Technologies Inc. The researchers presented the results at Eurospeech 2003 in Geneva, Switzerland, September 1 to 4. The research was funded by the Defense Advanced Research Projects Agency (DARPA).

Timeline: Now, 4 years
Funding: Government
TRN Categories: Applied Technology; Human-Computer Interaction
Story Type: News
Related Elements: Technical paper, "Speechalator: Two-Way Speech-To-Speech Translation on a Consumer PDA" posted at cmu.edu/~awb/papers/...speechalator.pdf," Eurospeech 2003 Geneva, Switzerland September 1-4

I can see it now... (4, Funny)

stienman (51024) | more than 10 years ago | (#7839911)

"Are you speaking the english?"

"I speak to the English, it's the Americans I won't talk to..."

-Adam

That's great, but ... (2, Insightful)

fastdecade (179638) | more than 10 years ago | (#7839914)

First can we have a PDA that does decent text-to-speech or speech-to-text, preferably both.

A hardware babelfish will revolutionise human communication later this century, but right now you need both of the above before you can begin to contemplate speech-to-speech. I can't imagine any serious algorithm at this time would attempt direct translation, without an intermediate text translation phase.

Bit OT: Considering the interest in E-Books, I don't know why music players and PDAs force users to download wave forms when we could just download text and convert using a cheap text-to-speech synth.

Re:That's great, but ... (2, Insightful)

cavebear42 (734821) | more than 10 years ago | (#7840063)

Forget PDA, I would like any software that can do a decent Speech-to-text. Every year of so I try all the latest stuff. Every year I keep typing. It is more likly that the rest of the world will learn english than we will have an effective translator in real time.

Re:That's great, but ... (1)

netsavior (627338) | more than 10 years ago | (#7840114)

Are you serious? an e-book read by stephen hawking? I download the mp3s because people reading books do accents and sometimes do fun voices, *sigh* it reminds me of kindergarten all over again. I could not listen to a synth for 80,000 words.

Re:That's great, but ... (1)

burns210 (572621) | more than 10 years ago | (#7840459)

ever use a mac? Honestly, i have been looking to get one, and their builtin speech recognition is fairly decent... much better(without any training, this was just a show model i walked up to and played with) than i would have expected, even after used popular speech programs and training them.

It is command based.... "Minimize all windows" "Close this application" and whatnot, but it is quite accurate(not perfect).

Maybe Jobs will have OS XI allow for plain-english input, rather than command based only :)

Yelling Helps (3, Funny)

aredubya74 (266988) | more than 10 years ago | (#7839921)

Outstanding. This thing will finally make the common Ugly American practice of yelling actually useful:

*hold PDA to face* Ahem! "WHERE IS THE BATHROOM?!" *hold PDA to foreigner's ear*

Re:Yelling Helps (0)

Anonymous Coward | more than 10 years ago | (#7840137)

Unfortunately, you might be shown a room that actually has a bath in it but no sign of a toilet ;)

Anyway I always thought that you Americans travelled with pistols and used those to communicate with non-english speakers.

Re:Yelling Helps (1)

aredubya74 (266988) | more than 10 years ago | (#7840327)

Unfortunately, you might be shown a room that actually has a bath in it but no sign of a toilet ;)

Obviously, you have never been to the dorm or apartment of an American college student. Toilet? Sink? Tub? Hey, whatever you can hit.

Anyway I always thought that you Americans travelled with pistols and used those to communicate with non-english speakers.

That's only NRA members and star athletes, and even then, they yell first, shoot later.

Obligatory AYB Reference (1)

Eberlin (570874) | more than 10 years ago | (#7839931)

...but since that's way too obvious, I'll leave it to the casual slashdotter to fill in the joke.

Let's face it, language butchery is funny. To do so automatically is so much more amusing! I mean I installed festival on my machine just so I could hear the synth voice say stuff like "beeeeyotch" and "retaaard" -- imagine how well you could offend in different dialects!

I suppose it does have legitimate uses...but what fun is that? Then again with the quality of translation software nowadays, it should be amusing nonetheless. If nothing else, maybe we can use it to come up with retranslated English to use as virus subjects. Maybe we could come up with gems better than "I send this to ask opinion for you. Don't show anyone!"

I wonder... (0, Redundant)

NeoThermic (732100) | more than 10 years ago | (#7839946)

If it can translate 'All Your Bases Are Belong To Us' correclty?

PDA speech translator... with ear bud? (1)

192939495969798999 (58312) | more than 10 years ago | (#7839978)

If it talks in and out, and uses an ear bud, it would be like being able to speak the language, albeit with a terrible accent, and occasionally offending the prime minister! That would be cool.

What did you say? (0)

Anonymous Coward | more than 10 years ago | (#7839990)

I do not know the location of the hovercraft of which you speak, but the eels sound potentially appetizing.

Takes the fun out (1)

plinius (714075) | more than 10 years ago | (#7839994)

The fun thing about travel is trying to undertstand people and them trying to understand you. Most people want to learn a little English and many Americans want to learn anything else (other than Spanish of course, which Mexicans have made them think is a peasant language). Anyway, with a machine I think it would be awkward, and it may make the local person feel a little inadequate (I got a PDA, you don't).

"My hovercraft is full of eels" (5, Funny)

NZheretic (23872) | more than 10 years ago | (#7840010)

With apologies to the python crew...

Text on screen: In 2004, the World Trade Center lay in ruins, and foreign nationalists frequented the streets - many of them Arabs (not the streets - the foreign nationals). Anyway, many of these Arabs went into tobacconist's shops to buy cigarettes....

A Arab tourist approaches the shopclerk. The tourist is talking haltingly into a PDA.

Arab: I will not buy this record, it is scratched.
Clerk: Sorry?
Arab: I will not buy this record, it is scratched.
Clerk: Uh, no, no, no. This is a tobacconist's.
Arab: Ah! I will not buy this *tobacconist's*, it is scratched.
Clerk: No, no, no, no. Tobacco...um...cigarettes (holds up a pack).
Arab: Ya! See-gar-ets! Ya! Uh...My hovercraft is full of eels.
Clerk: Sorry?
Arab: My hovercraft (pantomimes puffing a cigarette)...is full of eels (pretends to strike a match).
Clerk: Ahh, matches!
Arab: Ya! Ya! Ya! Ya! Do you waaaaant...do you waaaaaant...to come back to my place, bouncy bouncy?
Clerk: Here, I don't think you're using that thing right.
Arab: You great poof.
Clerk: That'll be six and six, please.
Arab: If I said you had a beautiful body, would you hold it against me? I...I am no longer infected.
Clerk: Uh, may I, uh...(takes PDA, talks to it)...Costs six and six...ah, here we are. (speaks weird Arabic-sounding words)
Arab punches the clerk.

Meanwhile, a cop on a quiet street cups his ear as if hearing a cry of distress. He sprints for many blocks and finally enters the tobacconist's.

Cop: What's up
Arab: Ah. You have beautiful thighs.
Cop: (looks down at himself) WHAT?!?
Clerk: He hit me!
Arab: Drop your panties, Sir William; I cannot wait 'til lunchtime. (points at clerk)
Cop: RIGHT!!! (drags Arab away by the arm)
Arab: (indignantly) My nipples explode with delight!

Just wait ten years (5, Interesting)

mschuyler (197441) | more than 10 years ago | (#7840041)

I believe PDAs are going to be tremendously transformed over the next few years.

1. Convergence is going to happen with a vengance. The Treo 600 is just the start. More and more apps will make it to the PDA. Speech recognition is one, and that sets up for another dybamic...

PDAs don't really need screens and keyboards if you can talk to them and they can talk to you. If they don't need those components, they can get a whole lot smaller. The next generation PDAs will be like a hearing aid, and the ones after that will be built into your glasses or an implant. That means less power, so less battery. Besides, it will be able to run on your body heat if not tap into your own body's electrical system, so it won't need a battery. Every improvemnt along these lines dwindles the size even more. A heads-up display, made transparent or opaque, ought to handle those times when you need to really observe rather than consult.

A combination of AI and connectivity will mean your PDA is your first line of defense in many of life's situations. Get pulled over by a cop and it will tell you what to do, what NOT to do, and contact your lawyer. Need a cop and it will call them and know just how long it's going to take to get there.

Medicine: It will have a complete medical history of you, remind you to take your meds, and monitor your blood pressure and other vita signs. If you have a heart attack it will call 911 with your location and be the first thing the medics consult when they get to you.

Personality: You'll be able to choose its level of humor and sarcasm. Although clearly a machine, people will develop meaningful relationships with them, at least they'll think so.

Connectivity: Everything you can think of, including your own house, which you'll call up to turn the heat up since you're coming home early. All teh Wi-Fi/cell connectivity you want will be built in.

Finances: It will know everything you do and provide access to your dough. If you get overdrawn it will be intentional because it will have real time access. It will have all the ATM/debit/credit stuff all on-hand. It will also be able to shop for you and tell you where the best deal is.

It will know all your friends and business associates and help remind you, "This is Joe. He's a Cougar. He knows you're a Husky, but don't rub it in. His kid just joined the Navy. He thinks LOTR sucks, and Rush is Right, so be careful. He drinks Guiness. His budget is 250K and he's looking to upgrade the Ciscos."

You'd never think of leaving home without this. Indeed, since it very well may be built-in, you won't have to worry about it. Just keep up the subscription.
'

Re:Just wait ten years (1)

RadioheadKid (461411) | more than 10 years ago | (#7840201)

Are you the same person that said to wait ten yearsf for all this stuff, ten years ago?

Re:Just wait ten years (1)

monkeyfinger (683580) | more than 10 years ago | (#7840205)

Convergence is going to happen with a vengance.

The ultimate in convergance will be when the PDA becomes part of the users body. Connecting with their nervous system and even their brain.

Re:Just wait ten years (1)

burns210 (572621) | more than 10 years ago | (#7840414)

I like your first point, that a computer needs input, output, and a processor. input and output don't have to be keyboard and screen, or a combo touchscreen as found in pdas.... voice recognition and sythesis(not necesarily anything CLOSE to true AI, just a well tuned bot) are just as good potentially for input/output...

The second part you mentioned, about medical history brings up a more interesting point... I don't see there being the all-mighty Palm Pilot to hold my life's history and be my google proxy when i get pulled over, i see many PDAs acting in a PAN(personal area network) to assist you.

A medical bracelet, instead of saying what you are allergic to, could hold a all your important medical history in a digital file, encrypted ofcourse.

Or a cam/eye glasses combo that feeds video into a Palm device, that might give feedback on say, face recognition(remember the guy you met last year at that party? ya, but what was his name?... your PDA could see his face, and match it against it's database)

I see many small devices working together, rather than 1 large device trying to do everything. Atleast that is what i would like to see happen.

The Russians developed a portable translator... (0)

Anonymous Coward | more than 10 years ago | (#7840057)

...with a small supercomputer stored in an over-sized novelty hat. You could only wear it for 10 minutes at a time before suffering permanent neck damage.

don't we need actual voice recog first (2, Insightful)

netsavior (627338) | more than 10 years ago | (#7840064)

my experience with voice recognition (yes even your beloved Via-Voice) is that it blows and will for some time. We probably need better speech recognition before we get speech to speech.

Save your money (0)

Swai (678004) | more than 10 years ago | (#7840068)

Instead of having all sorts of gadgets with you when traveling, plus all sorts of voltage adapters/convertors, here are some words that I think work fair good around the world.

For food:

  • Hamburger.
  • Mc Donald's.

For water:

  • Evian.
  • Perrier.

For car:

  • Toyota.
  • Nissan.

For a phone call:

  • AT&T.

If you need to send a message forget the bottle use the word FedEx.

After all is not that bad living in a world which brands everything but the air we breath :).

Never forget the useful SOS and the word NO. Is also helpful to learn the local word for please, potty, taxi and thanks.

Now all this words are for emergency situations else get a copy of the local travel guide from Mc Nally or any other source and you are more than set to go.

My final word is that we should'nt be umbilical cord attached to technologies left and right. Regards.

It might be faster and cheaper... (0)

Anonymous Coward | more than 10 years ago | (#7840105)

...for the United States to invade every country and impose the English language on each of them. We've been promised seamless voice recognition and translation for years and I don't see it happening anytime before Duke Nukem: Forever gets released.

Great! (0)

Anonymous Coward | more than 10 years ago | (#7840116)

Now if I can just get it to answer the question "So, what are you wearing?" then I can save myself from that $2.99/minute charge that keeps showing up on my phone bill!

History shows that no new technology really takes off until in becomes an effective distribution mechanism for porn...

Travelling (2, Informative)

elf-fire (715733) | more than 10 years ago | (#7840125)

Well. I have been to quite a few places where English was not exactly lingua franca. In most of these places semi-right pronounciation of foreign words would not have had a big impact. Hand gestures and my favourite dictionary (which contains pictures of just about anything one would ever need 'on the road') have always been sufficient to find a hotel, a train or bus ticket out and some food. For the latter: Just walking into a restaurant's kitchen and pointing at the visible ingredients (dead or alive ;) ) suffices, and can generate a lot of fun in the process :)

Better than a babelfish... (1)

jrumney (197329) | more than 10 years ago | (#7840136)

Not being content with translating humanoid speech, the Japanese have aimed their sights higher; dogs [takaratoys.co.jp] and cats [takaratoys.co.jp] . Cheaper than a PDA too, but they still need to work on the size and texture so it slithers nicely into the ear.

English speakers uncommon? (1)

El (94934) | more than 10 years ago | (#7840153)

I could see this as a boon to the tourist who travels to places where English speakers are uncommon. You mean like those horribly backward places that consider "aluminium" to be a 3-syllable word, and think "getting pissed" has something to do with being angry?

Re:English speakers uncommon? (1)

monkeyfinger (683580) | more than 10 years ago | (#7840258)

You mean like those horribly backward places that consider "aluminium" to be a 3-syllable word, and think "getting pissed" has something to do with being angry?

...and can't tell their arse from their fanny.

who needs a PDA... who needs a PDA... (0)

Anonymous Coward | more than 10 years ago | (#7840181)


Just don't forget your Protocol Droid

places where english speakers are uncommon (1)

dna42 (682340) | more than 10 years ago | (#7840197)

keep in mind that pdas are even more uncommon in those places, so i wouln't want to spend too much on such a device.
also keep in mind, that it actually is possible to learn a language, which does not happen to be the most widespread on this earth (or at least in those parts of this world you happen to travel to)...

As good as speech recognition even, but foreign? (0)

Anonymous Coward | more than 10 years ago | (#7840207)

Common, Lettuce divet a chalice and quite bean so whole of negativity!

Not as good as bablefish???? (1)

theolein (316044) | more than 10 years ago | (#7840211)

Babelfish is terrible at even translating Germanic and Latin languages and this thing is supposed to be worse than that?

I know that people want to solve everything with technology, but is it so much more difficult to learn another language or perhaps even a few phrases of the country where you are going to. Why does one even go to another country if one doesn't want to understand even the smallest part of that place?

Re:Not as good as bablefish???? (0)

Anonymous Coward | more than 10 years ago | (#7840256)

Obviously English is your second language, and the meaning of the word "a" in there went completely over your head. That and you have never read/heard/watched Hitchhikers Guide.

Bah! (2, Funny)

MoeMoe (659154) | more than 10 years ago | (#7840263)

Silly foreigner, don't you know everyone speaks English?

...where English speakers are uncommon (2, Funny)

JGag21 (678945) | more than 10 years ago | (#7840286)

Like Miami???

It even translates Muggese! (1)

Cordath (581672) | more than 10 years ago | (#7840317)

"Oh look honey! A local! I wonder what he wants. Use your iPaq to find out what he's saying!"

"Umm... He says 'Give me your iPaq or I will be forced to kill you and take your wife back to my yurt.'"

There are telephone translation services. (2, Interesting)

Moderation abuser (184013) | more than 10 years ago | (#7840465)

So all you need is a mobile phone. You phone up the number for the language you need translated to, tell the translator what you want to say and hand the phone over to the person you want to talk to. Quite expensive per minute, but cheaper than a PDA and very very handy in an emergency.

Course, you could learn another language, it isn't remotely as difficult as school makes it out to be. English is one of the more difficult languages to learn. If you learn, one of Italian, French, Spanish, Portugese you should be able to pick the others up fairly quickly. English is based on a Germanic language with a lot of the French and Roman influences chucked in on top, it's a real mishmash.

Does it cover Equus? (0)

Anonymous Coward | more than 10 years ago | (#7840483)

I want to talk to my mare!
Load More Comments
Slashdot Login

Need an Account?

Forgot your password?
or Connect with...

Don't worry, we never post anything without your permission.

Submission Text Formatting Tips

We support a small subset of HTML, namely these tags:

  • b
  • i
  • p
  • br
  • a
  • ol
  • ul
  • li
  • dl
  • dt
  • dd
  • em
  • strong
  • tt
  • blockquote
  • div
  • quote
  • ecode

"ecode" can be used for code snippets, for example:

<ecode>    while(1) { do_something(); } </ecode>