Slashdot Mirror

You can't just throw somebody into developing accessibility software

Very true! When I first started playing with Ubuntu accessibility, the blind were all upset about PulseAudio, but the Ubuntu screen reader worked perfectly for me. I didn't understand what they were upset about, and so I didn't put any time into tracking down the problems. Instead, I began training my ear to listen at very high speed. Now that I can hear properly, it's plain as day that a 1/2 second delay in speech was making Ubuntu as painful as a root canal. I've joined with other guys to track down the problems, and now Ubuntu is much better. Imagine debugging a sound problem you can't even hear? What if you worked on the speech synthesis tool I used to generate that sound file I linked to, and I came and said it sounds distorted today, compared to yesterday's release?

That's exactly what happened with this speech synthesizer. When you ask it to generate 22KHz sample-rate audio, it actually gives you 11KHz, which no one would notice unless they tried to play it at very high speed. I've reported the bug, but the developers have long since moved to a different company. I'm stuck with sucky 11KHz sound for life.

Combine that with Pixel Qi multi-touch displays and you've got a rock'in platform. This display does color video with the backlight on, but with it off it does E-Ink-like low-power black and white as good as any e-book reader. Power it with an Nvidia Tegra processor, and run Ubuntu Netbook Remix, and you've got one freaking awesome Internet tablet, with e-book reader being the killer app. Add a dynamically warping virtual multi-touch keyboard, and you've got a killer device. I think I'll need one of these babies for every member of my family.

Westlake, I don't blame you for not understanding the needs of the visually impaired. I certainly never did until I found my vision failing. In any case, the problem with audio-books is they are "professionally" read super sloooooooooow. About as slow as your mother read to you as a child. Here's a sample of how fast I listen. Smarter adults (like those of us here?) tend to read at 250 wpm - 450 wpm. I listen at 550 wpm. So, I'm currently reading The Half Blood Price, which takes under 8 hours with my TTS engine, at 550 wpm. The freaking audio-book is over 19 hours! I'd ... rather ... have ... teeth ... pulled. I'm not alone in the VI (visually impaired) community. J.K. Rowling's audio books are an insult to the blind, as are all author's works who do not release a text version readable with a freaking screen reader.

I agree, though, that it's hard to "punish" a billionaire with Pirate Bay. However, feel free to use the plight of the blind as your excuse as you pirate J.K. Rowling's works.

It's wonderful, but I doubt you're soon going to be replaced. The advantage of the computer generated voice (text-to-speech, or TTS), is that it's really fast. Here's a sample at around 450wpm.

I'm a programmer by trade, and a hacker/geek by choice. The TTS engine will enable me to remain highly productive even if I can't see. There's a blind guy getting a Ph.D. at NC State in computer science who has been giving me tips. He listens at around 850wpm, which is really amazing. He is probably just as good with a JAWS screen reader using the web as a person with normal vision, maybe better.

Anyway, for the elderly suffering from macular degeneration, there's probably not much use in learning speed listening. For those of us raising families and trying to be productive in the workforce, it's a godsend. Mostly, I'm listening to books rather than reading to train up my listening speed, preparing for the day I can no longer read. However, I have to admit, it's really fun listening to books that fast. I add the interpretation in my head, like I would if I were reading. The computer voice is fairly mechanical and flat.

Get any PDF, HTML, TXT, etc. Copy and paste it into gedit. Orca does a great job reading from gedit.

Creating the mp3 is trickier. Save the TXT file from gedit. Then, strip all the UTF-8 characters, like "circumflex", which is easy: just strip all the characters with the high bit set using a simple C program called stripUtf8. Then, use a customised version of the Voxin 'say' program to create the .wav, and 'lame -V2 file.wav' to create the mp3. To use the 'say' program, you'll need to pay under $10 to a non-profit in the UK. I hate to sound like a add for them, but that's the only legal way to get it cheap.

I've put the source for stripUtf8 and my customised 'say' program here.

Well, "words" are a bit fuzzy. Openoffice reports this text as 925 words. This is an mp3 of Orca reading it, which lasts 120 seconds. It's fun to listen to. I'm on my 7th novel in 4 weeks, which I play in the car, at the doctor's office, or anywhere else that's normally down time.

Well, "words" are a bit fuzzy. Openoffice reports this text as 925 words. This is an mp3 of Orca reading it, which lasts 120 seconds. It's fun to listen to. I'm on my 7th novel in 4 weeks, which I play in the car, at the doctor's office, or anywhere else that's normally down time.

Not a bad idea, but where would I publish it? I could post it on my Dumb Idea of the Day blog, but no one reads it (which is ok with me). I would certainly be interested in writing an article about coding for cache performance.

Yeah... I know, pretty vain. I got it shortly after Bill Clinton was no longer president. The site sucks, but I post to my dumb ideas sometimes.

Yeah... I know, pretty vain. I got it shortly after Bill Clinton was no longer president. The site sucks, but I post to my dumb ideas sometimes.

Here's my scary "Dumb Idea of the Day" In short, should I start an open-source license plate recognition project on sourceforge? It would likely encourage millions of home-based traffic snoopers world-wide... kinda scary

Sorry... :-P I would never promote my own stupid blog on slashdot... :-)

I think being open for development isn't enough, you also need to spend a couple $100 million or so in marketing...

There are several web sites out there that do nothing but publish your ideas (like these guys) so that they can be recorded as prior art. The stupid part is that they charge a very high fee. There needs to be a free site that is add-sponsored and community supported. In fact, if no one else does it, I'll do one. I already have my own dumb idea blog for this purpose. If a few of you respond to this post, suggesting that I actually provide this free service (and maybe some nice ideas about what web host, what CRM software, what to do to get paid adds, etc), I'll go ahead and do it. If any of you would rather do this yourself, please say so. I'd like to be an early user.

So, for example, with a post like this, will somebody in a dark suit and glasses show up at my door tomorrow?

Blasphemy #1: I've heard from a claimed friend of one of the inventors of RSA that it was cracked it years ago. Yet, it continues to get worldwide use. Sure my friend was probably full of it... but who am I suppose to trust here? The government?

Blasphemy #2: One of my close friend's mother had to switch fields from Numerics after she published some papers considered too sensitive. It had something to do with factoring.

Blasphemy #3: Anybody else notice that quantum computers have been proven to be capable of factoring really well, but no one has shown that they can solve any NP-hard algorithms? Come on... factoring isn't NP hard.

Then, there's just some silly stuff I've noticed about crypto. Why do we always seem to use encryption just a generation or so ahead of what is needed to crack it? SHA-1 for example... And, why do we encrypt one small block at a time. Each encrypted file usually gives many independent chances to crack the key, and in many cases, some of those blocks have known data. Also, public key is great, but secret key can be easily shown NP-hard to crack (in terms of secret key length) with semi-reasonable assumptions, while public key has no such simple proof. I personally have been trying to prove that no public key system can be NP-hard, but what the heck... I'm not that good. However, I do believe it's probably true.

It seems any time you start talking about crypto, you get assailed by experts telling you just how full of it you are. Consider something simple, like generation of random numbers. Just claiming you can do a good job brings nay-sayers out of the woodwork. See:

        http://linux.slashdot.org/comments.pl?sid=193904&c id=15899118
        http://www.billrocks.org/rng

for how to do it well. Any child could do it (well at least my geeky 6-year-old).

Everything about crypto is scary... Are we being manipulated into using weak encryption? Is there some invisible line, which if crossed, bad things can happen? The scary part is the unknown.

--

Just because your paranoid doesn't mean the world isn't out to get you.

For anyone interested, I'm developing a web-site to document my old RNG at http://www.billrocks.org/rng. Enjoy!

There's a great white-paper on Tesla Motor's site:

http://www.teslamotors.com/learn_more/white_papers .php?js_enabled=1

Basically, hydrogen sounds like it wont work, not in the near-term. On the other hand, new battery technologies are hopefully close at hand, like the ultra capacitors from EEStor. If we get this working, it could mostly eliminate our oil imports, and clean up the air. I posted a blog entry at:

http://www.billrocks.org/ideas/index.php?/archives /15-Saving-the-World-with-Electric-Cars.html