julesjones

Friday night a week and a bit ago I felt up to experimenting with using Dragon to write new fiction material. I specify fiction, because a lot of my blog posts over the last few months have been written at least partly with Dragon, and an awful lot of my typing at work has been with Dragon rather than keyboard. But fiction's different, for various reasons.

The current primary WIP is one that I wouldn't even attempt to use Dragon on. There's too much staring into space while I turn shape of story into words, and dealing with the error correction would thoroughly disrupt my train of thought. But occasionally I have a chunk of text right there in my head, and that might be more amenable to scribbling it down quickly via dictation rather than via keyboard or handwriting recognition (I still carry around a Palm IIIxe for the latter purpose, even if it doesn't get much use nowadays).

That still wouldn't work for anything longer than one or two sentences using my normal mode of dictating, which is to dictate a sentence or two and then do the corrections. But Dragon can also transcribe from a sound file with previously recorded dictation, and my shiny new headset came with sound recording software.

I had something that wasn't full on "the text is in my head and I simply need to type it before I forget the words", but did have a reasonable chance of me being able to witter on for more than one sentence before having to stop to formulate the next bit. That makes it feasible to dictate a passage, and then go through the complete transcription file afterwards to clean up the errors by listening to the recording and making corrections. (Cleaning up the errors via Dragon is essential to improve recognition in the future.)

Alas, I'd left it a bit too long before doing something with it, and some of the words had escaped. There was much pausing to think of how the next bit should be phrased, and thus many points where I paused the recording in order to stare into space.

The actual dictating into recorder bit -- not entirely a success. It made me much more consciously aware of how much I sit and spare into space, for one thing. :-) But the two big problems I consciously noticed were that I find it much more difficult to compose the next sentence without the visual feedback of the previous sentence on the screen/page in front of me, and that I still find it disruptive to my train of thought to have to explicitly dictate punctuation. However, the punctuation problem is, I think, less noticeable when doing this sort of "dictate for later transcription" than when doing real-time dictating/correcting.

The sound file I ended up with is 9m38s long, according to Audio Commander. That's with a fair bit of hitting the pause button when I needed to stare into space, and I'm not quite sure how long I was actually dictating for, although I doubt it was longer than 20m.

The raw transcript has a word count of 829 words. Many of those words are not what I actually said. Some of them are words that I did in fact say, but intended to be dictation commands rather than text. The transcription is actually pretty appalling, compare to how I'm getting on with using Dragon at work to dictate letters etc. :-) This may be partly because I didn't remember to RTFM before starting this lark, and thus didn't set up the digital recorder as a separate source. I did shift the accuracy-speed balance all the way over to accuracy, so that's not the problem. Or it may be that I haven't written any fiction with this profile, so its sampling of my likely pattern of word usage is off.

It's taken about 20 minutes to go through the transcript playing back the audio and correcting it where necessary. That's just the first pass, to correct the transcript to what I actually said, glitches and all. This is necessary to teach Dragon to adapt to my speech patterns -- if I don't correct it accurately to exactly what I said, rather than what I meant, that will decrease its recognition accuracy in future.

Next pass will be to go through and take out the commands it mis-recognised as text, and do the corrections I was trying to do with those commands (typically deleting a few words because I'd changed my mind about what I wanted to type). And after that I'll do a revision pass.

So, slow and tedious. But doable, and could be useful in certain circumstances. I might put the raw transcript and the various editing passes online, if people are interested in seeing what sort of errors Dragon is making.

I'm actually up to trying to type 300 words this weekend, so did so. It ended up as 400 words. :-) Taxman has gone from 27977 to 28375 words.

I've also got a possible market and some further editing suggestions for Circle of Glass, which will probably have to wait until Alex and I are available and alert at the same time. Might poke at that tomorrow, if not it's probably going to be next weekend before we can do so.

I want to try using the dictaphone function in Dragon in combination with the sound recorder in the headphones software to experiment with dictating fiction. As a general rule, I can't do this, because various things about using Dragon break my train of thought and I lose track of what I wanted to write. However, I've considered in the past using a dictaphone and then feeding into Dragon, rather than correcting recognition errors on the fly.

I never got around to it, in part because it would still probably be rather disruptive to my train of thought as far as actual editing/rewriting on the fly was concerned. I was also rather reluctant to shell out on a digital voice recorder given that they were fairly expensive at the time, at least for one good enough to use with older versions of Dragon. However, the reason I was considering it was because I get the occasional bout of having the story pretty much ready to go in my head before I ever lay fingers on keys -- and *that*, even if it was simply a chapter or scene, would be much more amenable to dictate first and correct recognition errors after.

The muse dumped a story idea on me a couple of weeks back, along with a chunk of dialogue. Some of it's going to be sitting at the keyboard staring into space, but some of it at least I could probably rattle off into a dictaphone. And the sound recorder is a good way to see if this actually works for me. Unfortunately I've been too tired since then to be able to try, because composing my thoughts and then saying them accurately requires concentration. So that's probably tomorrow's writing task, if I get a decent night's sleep tonight.

For anyone UK-based thinking of buying Dragon 11.5, the UK online shop is having a six hour sale from 9 to 3 today on the Windows Premium edition, in basic, bluetooth and digital recorder versions -- more details here. From past experience with Nuance sales, do check the prices against the discount prices available from places like Amazon, because they're not always that much of a deal. But sometimes they do offer a worthwhile discount off street price, and these prices are a serious discount compared to what Amazon is offering me today.

I don't particularly recommend the Bluetooth edition. The included Bluetooth headset appears to be the cheapest, crappiest one of the Plantronics Calisto range, and I suspect an OEM-only model. The actual mike is fine, but it's an over-the-ear set with a one-size-doesn't-fit-everybody earplug and loop, which I find is impossible to keep in place for any length of time. There's no charging stand, only a flimsy charging cable. Nice idea to not have to be taking the headset off and on to move away from the computer, but if it doesn't fit your ear you're going to be constantly fiddling with it anyway. It's usable, it's useful for work where you don't want to be tethered to the computer, and it's useful for travelling because of the small size, but I'm about to buy a headband mike (specifically the model I've been supplied with at work, which seems to suit me).

and you can see from the timestamps on the previous two public posts just how long it takes to do anything with Dragon ....

I'm still working my way through the manual.Yesterday's discovery -- that saying "mouse click" does exactly that. Now, if you can more or less shove the mouse around the mat but clicking it hurts, this is extremely useful all by itself. But mention was made of the Mouse Grid. I left that for today. This command does indeed lay a 3 x 3 grid over the screen, you pick the numbered cell that you want, and a smaller grid pops up in that cell, with the mouse pointer in the centre. You can repeat this several times, and use the "mouse click" command as soon as you get the mouse pointer where you need it on the screen. This can be unbelievably tedious, but is also very useful when the verbal commands for selecting links fail you. I could have done with knowing about this a week ago, when even pushing the mouse around hurt.

And yes, there is a command for right clicking, and you can instruct it to press specific keys on the keypad. Of course, its recognition of individual letters isn't always that wonderful, but it does recognise the alpha bravo codes, which I am going to have to relearn.

And this little lot took me over 10 minutes to dictate. This is one of the reasons why it's not terribly practical for writing a novel with, at least for me. I await with interest Charlie Stross's report on his experiment with dictating to his iPad. (And I didn't have to spell out Charlie's name. Any of it.)

So, the error message I get when I try to open a PDF with Dragon running is:
"COM returned an unexpected error code: Details are HOOKERR_NONOTIFYWINDOW."
and Dragon then falls over, either immediately or within the next few minutes. Putting "dragon 10 acrobat reader" into Google got me the following as the first result (note that it's affecting Dragon v11 as well):

http://acrobatusers.com/forum/accessibility/adobe-reader-x-interface-problem-dragon-naturally-speaking-11/

And in case that thread disappears, Dragon and Adobe are blaming each other, the possible fixes are to buy the full Acrobat X, turn off protected mode in Acrobat Reader (works for some people but not others), or roll Reader back to a previous version. Or, as the original poster in the thread was thinking of, finding some other pdf reader...

The second link was
http://www.speechcomputing.com/node/4025
and apparently it was all over the speech recognition fora when Acrobat Reader X came out. There's a link in that thread to a technical note at Nuance (the people who own Dragon):
http://nuance.custhelp.com/app/answers/detail/a_id/6224/kw/Adobe%20reader%20X

If it comes down to having to ditch one of the programs, it's Reader that's getting the boot. And I'm not impressed with Adobe breaking one of the most widely used accessibility programs out there while supposedly improving accessibility.

ETA: another useful link: http://www.speakeasysolutions.com/blog/2011/03/07/dragon-error-com-returned-an-unexpected-error-code-details-are-hookerr_nonotifywindow/

The RSI is now sufficiently bad that Dragon is no longer an option. I've spent a lot of the last two days relearning how to use it. When I say "a lot", I mean 15 minutes at a time, and then taking a break. As it points out somewhere in the manual, you can also get RSI of the vocal cords. And it's a good thing that one of the commands I could remember is the one to turn the microphone off for a bit, because with just having got over a sore throat, I need to stop to clear my throat every sentence or two. It's also getting bloody annoying that the microphone doesn't fit properly and keeps drifting out of alignment. It doesn't help that I've acquired reading glasses since I was last using this program regularly.