"Dilbert" Creator Gets Voice Back

Scott Adams lost his voice 18 months ago to a disorder called Spasmodic Dysphonia. One day, it returned. He is apparently the first person in history to recover from this malady. Read his account. It is inspirational. I can't find any other word for it.

Elaborate ruse? Maybe not...

[Shoeler]

Fellow Dilbertites,

It seems the great overloard Adams was in fact inflicted by the great malady. [nih.gov] Rejoice at his miraculous recovery!

PS - I was quite confused at first as to the authenticity of this until I got goog-learned [google.com]. It seems it really does exist [emedicine.com], he very well may have had it [typepad.com], and if he recovered was indeed a miracle. However, it could also be an elaborate ruse, as I would expect from a satirist of his pedigree. :)

Re:ffs

[AKAImBatman]

Gah, beat me to it. Ah well, here's what I was going to post (slightly more polite):

KDawson, I just thought you'd like to know that the Enlightenment category is for the X11 Window Manager [enlightenment.org] by that name, and not "enlightening" topics. Unfortunately, Slashdot doesn't really have an "Inspirational" category. About the best you can do is "Entertainment" and "Links". Since this is the third time [slashdot.org] you've been in want of an inspirational category, you might consider talking with Taco about remedying the situation.

Re:ffs

[Shoeler]

Do none of you find any irony in the choosing of a double-entendre topic / icon and a condition like this???

Spasmodic Dysphonia

[Dan Slotman]

Wikipedia [wikipedia.org] has a nice article on Spasmodic Dysphonia.

As the blog indicates, this is thought to be a neurological condition. When I was studying AI as an undergrad, we learned a lot about neural networks [wikipedia.org]. This seems like the sort of thing that could happen if the brain's speech area's neurons somehow became trained to stop delivering impulses for "normal" speech. In this case, it would be theoretically possible to train the network back to normal levels. Of course, it could be something completely different.

Here's wishing Scott the best.

Enlightened

[kdawson]

Thanks, I changed this. There really is no perfectly appropriate topic for this story.

Re:ffs

[dan828]

Seriously, I've been reading slashdot for a couple of years (but not TFAs of course), and I didn't even know that there was an enlightment category, let alone not even recognizing the icon when I saw it. Though it should have been used on the "Yellow Dog Linux on the PS3" story from a while back http://linux.slashdot.org/article.pl?sid=06/10/16/ 1342243 [slashdot.org] , because they are apparently going to use E17 when/if it comes out.

Re:I met a guy with that once

[tehshen]

Know what's crazy? I can't talk properly either, yet when I sing, I am fine (despite singing badly).

There are a bunch of reasons that I've heard for this: that the words are longer so it's harder for me to mess them up, something about music and talking being in opposite hemispheres of the brain, and something about the singing voice being smoother or calmer than talking.

There was a story a while back about some girl getting a speaking aid where whatever she says is "echoed" into her ear, giving the impression that she's talking with someone else, which makes talking a lot easier. Yeah, here it is [bbc.co.uk].

Hooray to you, mr Adams. Us silent folk aren't all bad.

Re:There's a problem though

[cnettel]

Ok, but if we put it like this: a general multi-layer perceptron, which is often what "neural network" means in practice in an AI context, is quite dissimilar to a real neural network. You can't even get a feedback! (Which is kind of logical, since a MLP generally doesn't model time per se.) Back-propagation training is also quite different from the self-promoting mechanisms that are now believed to be significant for selection of neural connections.

There are some similarities, and it's certainly possible to model biological neurons and systems in a machine. Those models will bear some similarities to neural networks used in classifying tasks, but there are also similarities to a whole range of (other) graph problems. It's kind of like the relation between ray-tracing and triangle/Z-buffer based rendering. The latter is a way to approximate the former, sort of. They have some similarities, and programmable hardware that's good for doing the latter might be tweaked to do the former as well, but you don't get a raytracer just by cranking up the polygon count, as the whole strength of the normal rendering paradigm is based on greatly simplifying assumptions that are centered on Getting Stuff Done.

Re:I suffered a similar problem

[SEAL]

Just as a follow-on, for those curious:

My condition was not spasmodic dysphonia. It was classified as injury-induced dysarthria [wikipedia.org].

Diane Rehm has struggled for years with this

[gelfling]

It may be incurable but it's not unmanageable. see http://wamu.org/programs/dr/diane_rehm/ [wamu.org]

not the first to recover speech

[stupidsocialscientis]

Diane Rehm, the NPR interviewer has this disorder and does 1-2 hours of show every weekday. scott is hardly the first to recover function, unless he means completely typical function (i couldn't decide which he meant after RTFA.)i hope for his sake it is a permanent, full recovery.

Re:Elaborate ruse? Maybe not...

[steveha]

...it could also be an elaborate ruse, as I would expect from a satirist of his pedigree.

It is ironic that you say this, because he wrote an elaborate short essay about this topic. The first blog entry where he announced his malady was here:

http://dilbertblog.typepad.com/the_dilbert_blog/20 05/12/the_problem_wit.html [typepad.com]

A quote:

It's bad enough to find out that I'll probably never speak normally to another person for the rest of my life. But to make things worse, my notorious cleverness makes people think I'm joking when I explain it.

steveha