Voice recognition: A Status Report

Siri, Dragon, Samsung, Microsoft Windows, … , the list goes on and on. Nearly everyone in the computer device game recognizes that as the hardware get smaller and smaller—and their onscreen keyboards shrink proportionately—more reliable voice recognition will be a game changer. It has to be done: In the near future, smart phones—or a part of them anyway—will shrink again and be strapped to your wrist ala Dick Tracy. Google Glass, in which the display is projected on your glasses, is perhaps the ultimate example of shrinking displays, at least until contact lenses become computer displays.

In a quite different domain, coming as soon as the next decade or so, many new kinds of devices will call out for voice recognition—perhaps literally. Think of your automobile, your dishwasher, the lights in your home, things like that.

But the bottom line today is that most voice recognition is crappy. Even Microsoft's own speech recognition engineers admit its failings. Speech recognition at Microsoft now has in fact been relegated to the Ease of Use section of Windows Control Panel, downplaying it and implying, without quite saying so, that it is aimed at people unable to use a keyboard. Yet Willi who is not all handicapped—unless one chats with his wife—has dictated literally hundreds of thousands of words, in some cases complete books. He hasn't used a keyboard for years. He is dictating this blog post using Microsoft's speech recognition, which is considerably better than you might expect, both to give commands and to dictate words, though it still has a way to go. He has also used both Siri and Dragon to some extent (enough to know that they're not much good—yet).

As to speed, there's no question in Wili's mind that he can dictate faster, using Microsoft's technology, than the fastest typist, assuming there is a prepared text for both, and that includes corrections. But that's not really the point; when dictating without a prepared text, one can only speak as fast as one can think, and for most of us that's not very fast. Another factor in speech dictation that is often overlooked is that what one writes is often more natural and smooth than what one types on a keyboard. This can be important for writers.

To get one's head around the current status of speech recognition it is important to understand a few basics:

First, the distinction between speaking commands and dictation. Commands are far easier to understand for the simple reason that there are a limited number of them from which to pick, so that when you speak, it is a relatively simple task to determine which you mean; speech recognition is a statistical game. It is a pretty simple job to discriminate between "Lights on" or "Lights off".

Dictation is a far more difficult animal because when you say something the realm of possibilities includes every word in the language you're using (and yes you can use languages other than English; it's a global world now in the computer business). In the English language there are, give or take, 200,000 words. There are more, really, if one counts medical terms and other scientific esoterica. And then, with dictation there's the added difficulty of distinguishing what you mean when you say a homonym such as "to", which could be interpreted as "to", "too", or "two ", depending on the context. So you see this is not a simple proposition. Yet surprising progress has been made in this direction.

The next consideration is ambient noise: speech recognition in a quiet room is vastly easier to understand than speech in a work cubicle, an automobile, on the street, or in a bar—in roughly that order of difficulty, unless you inhabit quiet bars, as does Willi when he is able.

Finally, the quality of the microphone that is trying to listen to what you're saying is often important, depending upon the other three considerations listed above. Willi, at this moment, is in a quiet room and is being listened to by the simple but profound little microphone built into the edge of his computer a Microsoft Surface Pro. Yes, there are noise canceling microphones; Willi has tried them; they don't help a lot.

These are the basic parameters with which to evaluate the practicalities of speech recognition. To sum up the current status of speech recognition one should take away the thought that more work is needed both on the software, which is a complicated matter in itself, but even more work is needed on some sort of noise filtering hardware. This remains the single biggest roadblock to practical, everyday use of this technology.

One does not always have a quiet room to work in. And it is "on the go" when one most needs this technology. Willi thought about a throat MIC, as was used in airplanes for WWII where the noise from the engines was overwhelming. Perhaps a new such device could be made as cool-looking throat ornamentation—a leather choker encrusted with fake diamonds. It would take sound directly from your larynx. Never underestimate the benefits of fashion to sell an idea.

Feast or Famine

FUQ Department

Those of us of any age whatsoever tend to think of the finances of our nation as similar to the finances of our families: if you're running out of money, stop spending so much.

This seems to be the mindset of the Europeans right now; they are all working on austerity plans. The entire EU, though some members more than others, have zipped their wallets shut tightly. The Chancellor of the Exchequer, in London, is reining in, as is the Chancellor of Germany, and the Nordic countries have frozen their budgets. But the Latin nations in the south are not so sure about this deal. Greece and Italy have just had an election and the people, fed up now with austerity, threw out their tightwad governments which, they claimed, were listening too much to the tightwad Germans who were whining at being stuck with the bill after the southerners had dined sumptuously.

Here in America, we can't seem to make up our minds. The Democrats want to spend while the Republicans want to cut the budget. Thus we come to our Frequently Unanswered Question: Is it better for the government to quit spending so much or is it better to have them prime the pump once again with a few trillion which we don't have but which we can print just as fast as those presses will run, and that's pretty fast.

Paul Krugman, a Nobel laureate for Economics, says we should spend more. On the other hand President Obama got a Nobel prize for peace before he'd hardly got down to work. So who knows about this Nobel business? Anyway, Krugman claims that the national economy is not like your family's budget, and he explains it pretty well this way: unlike a family, for a nation of people, one person's spending is another person's income; as a consequence we can keep printing money until inflation starts to go up, which it hasn't yet, much. Still, the Europeans have economists too, and Willi bets at least one of them have a Nobel prize, probably a flock of them do. Yet they seem to come to the opposite conclusion: spend less!

It seems to Willi incredible that economists who study this matter cannot come to a consensus. Once again, Harry Truman was right on point when he said something like this: Give me a one-armed economist any time, that way he can't say, "On the one hand…"

We should by this time have enough computers in the cloud to simulate just about anything and come up with the answer. We have an IBM computer, named Watson, which can beat Ken Jennings at Jeopardy, and one named Deep Blue which can make grandmaster chess players quiver. There is a helluva rig over there in Europe on the border of France and Switzerland called the Large Hadron Collider which has analyzed some billion trillion particle collisions in order to find a boson named Higgs. They think they've spotted him, but now they think he might have his whole family with him—though maybe they're just angling for some more money. The point is these people can do a lot of shit. But they can't seem to come up with the answer to a simple yes/no question: spend more/spend less. Why is this so hard?

Willi doesn't know.

The FUQ Department

Willi has decided on a change, or addition, to his blog, modernizing it so to speak. The first thing to say about this new department is that it is not what you think it is—those of you of scurrilous mind—you know you are. No, its meaning is simply this: the Electronic Media is full of topics intended to assist you in finding out how to use some arcane bit of software or a hardware device, something for which there are absolutely no intelligible instructions and people are forced to ask, if they're able to find a path to do so, just how to do something or other. These are then published by the developers or manufacturers as a sort of poor man's instruction booklet. These questions have quite naturally come to be known as Frequently Asked Questions, or FAQ's.

Willi's new FUQ department is similar in a manner of speaking, but different: it is concerned with Frequently Unasked Questions, or FUQ's. Sometimes it can stand for Frequently Unanswered Questions. It seems to him that FUQ's are in many ways more important than FAQ's, more subtle perhaps, yet more important for all of that.

This wide ranging and far reaching department will attempt to expose the numerous deficiencies that he finds to be slowly eating away at good and useful reportage—or anything else that the department can plunder for useful purposes. He supposes that it might have been called The Curmudgeon Department; he is after all 79 years old and one ought to expect these things at his age, but somehow that title didn't have as much zing to it, and zing is an important thing.

Stay tuned.

Rope a Pope

The media coverage of the election of a new pope at the Vatican is incredible. Willi had thought that this was the secular age. Wrong, wrong, wrong! Perhaps it's only a slow news cycle, but people are fascinated with this 2000 year old process. It makes him wonder what the hell is going on.

The New York Times, not exactly known as a bastion of religiosity, cannot get enough of this story, they've published literally dozens of articles. From the merest rumors of the machinations of the Red Hats, in conclave, to the precise details of just how black smoke or white smoke is generated after each round of balloting—here's a hint: the holy spirit doesn't do it—these people are fascinated. Why?

Pope Francis is now the head of the Church until he dies—or decides to retire. Can there be any one now in Europe, Asia, or the Western hemisphere, who does not know that name? Willi asks: what is the name of the top guy in the Eastern Orthodox Church? Who is the Large Lutheran? The Big Buddhist? The Rebbe Rabbi? The Hefty Hindu? The Major Muslim? Vastest Voodooist? No, you can't name any of them. But Willi bets that you can name the Pope. Why?

He doesn't know.

What is Surface Pro?

Microsoft has come out with two new computers, the Surface RT and the Surface Pro. The RT was released last fall and in terms of price and functionality it seems intended to compete with Apple's iPad. The Surface Pro, released in February, is quite a different animal. While they look nearly identical from the outside (the Surface Pro is thicker and is heavier—2 pounds) they are very different inside. The rest of this post concerns the Surface Pro. First, the physical specs:
It has a 10.6 inch screen with a 16:9 form factor—as though designed to show panoramic movies. The screen resolution is very good (1920 X 1080); text looks clear and sharp. Inside, it has 4 Gb of ram (no options), and either 64 or 128 Gb of permanent storage in the form of flash memory, although the operating system and various odds and ends take up more than half of this space. It has an Intel i5 processor, about as fast as it gets in this genre (about four times faster than the iPad). It has a battery life of about 5 hours. It uses Wi-Fi connectivity, and it also has Bluetooth, but it does not yet have cellular.
There are two available keyboards which double as snap on screen covers, one is a kind of a thin key touch pad and the other somewhat thicker one has real keys with a certain amount of throw to them. It also comes with several interesting varieties of onscreen keyboards with different arrangements and even keys for different languages. It has a cool little kick stand so that you can prop it up on a table or desk (but probably not on your lap). Oh, and one thing more, everything responds to your touch, and it comes with a stylus too so that you can write in certain apps and draw pictures too—if you are able.
It is in the operating system software that the Surface Pro is quite different than the RT, and this can at first be confusing to old time Windows users. There are really two distinct parts to this operating system. The first part, one could call it the Start part, is nearly identical to the RT: it has colorful tiles with which to execute certain functions of the computer. These tiles are invariably what one sees in advertisements for the Surface computers and that is certainly because it makes a pretty montage, one so different from the classical Windows operating system that one begins to think they've changed everything around. But now we will see the difference between the Surface Pro and the Surface RT: one of these tiles is labeled "Desktop", and tapping this tile whisks you away to the classical Windows where you will feel quite at home—it will seem as though you had never left; what's with this? Well here is the secret:
When you get to the desktop and you want to do something, you will notice that there's no Start Button! What has happened here is that the so-called Start Screen, that cute set of tiles, is the equivalent of the Start Button though with considerably more versatility and good looks. At least that is the way that I think of it. That "light bulb" thought cleared up something fundamental in my head, and if you are an old Windows guy, as am I, this thought model will probably work for you too.
After several days of reading reviews, written just as it was being released, I bought one of the 64 GB versions because 1) I could get it—they were sold out of the 128 GB ones—and 2) most of my data will be kept in the cloud anyway so I won't need a lot of space in the computer itself.
The main reason I like the Surface Pro is that I've been using tablets for years, even after our IT people thought I was a little nuts. And the reason they thought that is because I've never had a keyboard, onscreen or off; I just use speech recognition and a stylus. Now, in my old age I've become an amateur writer. I've written hundreds of thousands of words while never touching a keyboard. I was a little nonplused when several years ago Microsoft moved speech recognition into the "Ease of Use" section of Control Panel; I'm not handicapped, I thought to myself, but then I talked to my wife…
Whether you will like it or not is much harder to answer because it all depends upon just what you plan to be doing with it. Of all the reviews I've read, the majority simply assumed you would be using a keyboard. Many commented "Why not just buy a laptop?" Indeed, if you do a lot of keying that makes sense. But most of the reviewers were discombobulated because the machine seems to be neither fish nor fowl. It's a tablet with which you can run all your Windows programs.
I like to think that in a subtle way Microsoft is moving us along toward a future of touch and speech. It ought to be said that speech recognition still has some way to go. I find it acceptable because I've taken the trouble to learn its idiosyncrasies—and it has taken the trouble to learn mine. There are commands that must be used, and one has to memorize them to be efficient, and to some extent these commands change as one moves between apps. At first this can be frustrating. But as you go on it learns things about you as well. It's a two way street.
First off, you read to it and it learns how you pronounce things (are you from New England or from Jamaica). This will take 10 minutes or more depending on how much time you wish to invest in it—the longer the better. Then it asks you very politely whether or not it can look at the text that you dictate, such as your emails and your Word documents, and if you agree it checks them out to see the words that you use frequently. It took some time for it to understand how to spell my last name, Goetsch, but now after nosing around a bit it understands it gets it easily.
I have dictated this blog post using only the microphone embedded in the edge of the computer. This surprised me considerably since I couldn't do that reliably with my previous computer. I have though ordered a small speech dictation microphone that plugs directly into the microphone/speaker connection in the edge of the computer. That cuts down on the corrections.

Whatever happened to the paragraph?

Remember when a paragraph was a set of sentences expected to express some thought or notion that an author wanted to get across? In those long ago days even a single sentence could ramble on quite fluidly, commas here, semi colons there, and one might even encounter the occasional colon. I always liked the colon, perhaps because so scarce they seemed exotic.

If one goes back further still—yet not all that far—an extravagant paragraph might go on for a page or more, sentence after rambling sentence. This extreme could be annoying because if you were interrupted while digesting this big mouthful it was hard to find your place again when you got back to it.

When I studied grammar in high school, our teacher said definitively that a paragraph should express a complete idea. That mystified me; almost anything, even a word, can put across an idea: sweet, noisy, and I could go on. So that definition always seemed to me pretty shaky, worse than useless actually, a waste of time; I was no better off than before. But as the years have flown past, paragraphs have become shorter and shorter.

Now I think, the ultimate has been reached. I just read an article in the New York Times—which organization is, or was, staffed with echelons of editors. A writer could hardly get an article through this phalanx without having punctuation and organization checked upside down and backwards; and that doesn't even count fact-checking and other such editorial function. So people think of them as exemplars of this sort of esoterica.

Here is the article I read, not very interesting I found, after I got into it; I can give you the whole article in one sentence:

Profligate Greece spent more money than she had and now wants to borrow more, but the Euro people, tightwads, don't want to give her any; so she threatens to borrow from the IMF, but The Euros don't like that idea either because it makes them look cheap and unsteady.

Think of all the trees I might have saved. Why do I read articles like this in the first place? The answer is simple, if not edifying; the headline sucked me in: "Debt Problems Chip Away at Fortress Europe" I like to read articles that confirm my biases about the snooty Europeans. There you have it; not pretty, but human. But that's not my point.

Why the New York Times thought that this might be news—something that is new—I can't even imagine. Nevertheless the story, as I hypnotically read it, was mesmerizing; it was as though I were watching a public execution; the entire article consisted of about 31 paragraphs—and 33 sentences. Paragraphs! Sentences! Nearly every paragraph, if such they can be called, consisted of one sentence. One sentence! Consider this so-called paragraph:

    But that may be easier said than done.

Or this one:

The speculator's bet is a simple one.

That's right, these are entire paragraphs, brought to you by the Doyen of literary organization and punctuation. So, that's it for the paragraph. Muerte.

Why I am an American

If, after reading this title, you suppose that this is a patriotic screed you will be excused. You will be wrong, but you will be excused. If I read a title: "Why I am a Christian", or "Why I am a Lesbian", (neither of which would be true) I too would expect some sort of argumentation to follow. It's natural. But what I will write about here is nothing like that. This post concerns only linguistics. Think about it: Both 'Lesbian' and 'Christian' are qualities, while 'America' and 'Lesbos' are places.

I want to explain why we here in the United States have arrogantly grasped for ourselves the quality of being American, when certainly Argentineans, Canadians and all the many peoples in between on this side of our planet would seem to have an equal claim on the term. When traveling to other countries I used to be reluctant to say yes when asked "Are you American?". Often I would just reply "Yes, I'm from the States." I thought it sounded more humble, and I like that quality. But on reflection it is easy to see that Mexico consists of united states, just as we do, so that's not a very good answer either; what makes us think we are The States? So the thing is, there's just no good way around this problem. I have quite given up, and now just reply "Yes, I'm American." Let there be an end to this humility.

The forming of adjectives from nouns is actually a trickier business than one might at first think. If I am from Australia I have a pretty easy job: tack an 'n' on the end and you're done. Simple as pie. You're Australian; easy. If you're from Romania, ditto. Costa Rica, ditto, you are Costa Rican.

Now give a moment's thought to Canada. For some unknown reason they don't follow the pattern. They're not 'Canadan', like every other country name that ends in a, they are Canadian, they have, it seems to me, and somewhat pretentiously, added an extra fillip to the standard formula, which goes to show you that even the seemingly humble Canadians are not entirely without a sense of extravagance—perhaps the French influence.

If one hails from Britain, one becomes British, in the same way that an imp, having a good day, might become impish, or a ripening apple might become reddish, a state of being that implies, hints at a condition, not definitely stating it. The English are rightly credited with understatement. And further, this reticence is often carried to the further extreme of simple truncation as in, "Yes, I'm a Brit." And what are we to make of Japanese, Chinese, and even Taiwanese and Lebanese? The linguistic mind boggles. But the real trouble is this:

What can we call someone from the United States? United Statesian? Doesn't exactly roll around on the tongue does it? USian? I don't think so. How do you pronounce it? So you see the problem.

We needed a term and we took it. Get over it. And of all the places in the western hemisphere, the continents of the Americas, that might have a claim on the term—whether one likes what we do, or dislikes what we do—it would be hard to argue that we're not The Americans. Which will certainly put a shiver down the spines of Canadians and Europeans (who by rights ought to be Europan), many of whom speak pejoratively of American Exceptionalism. What they may not realize is that by assigning us this judgmental phrase they simply confirm the term we have assumed, American, since everyone knows who they're talking about.