I love my helper dragon.
As soon as I had discovered my sight loss I thought about what assistive technology could do for me.
There is a free iPad App which is called Dragon dictation. I downloaded and tried it and found it remarkably good. The first few posts on this blog were dictated using it. They needed a bit of clean up but basically the technology was an enormous help. I have never been a very good typist and the loss of ability to scan lines of text meant typing was now a slow and painful experience. The dragon solved this problem and even the free version was so good that I wished I’d been using it all the time.
But now I’ve acquired the full version, running on a desktop computer. It is far more powerful than the free version, as you’d expect, and what it can do is extremely impressive.
For a start the full-blown software can be trained for one’s voice and accent. It recognises a far wider range of spoken punctuation than the iPad software and also responds to commands. For example, if I say “Read the document”, I can listen to the document read aloud.
It can cope with alternative spellings such as “to”, “too”, “two”. You just ask it to “Spell that” and then choose from a pop-up list of alternatives by say which number item to choose.
It can even perform mouse actions with commands such as “Move mouse” and “Mouse click”.
When you’ve finished or want a break you just say “Microphone off” and sure enough the microphone gets switched off.
And it integrates with other programmes. Whatever I’m working with my helper is ready to take dictation.
All this will be familiar to anybody who uses this or similar software, but to me it was a revelation. I’m still very far from mastering it but I’m making good progress.
There are some downsides. One pitfall is forgetting to turn the microphone off. The other evening I left it on while I was listening to Radio Four and it took down about 1000 words of inaccurately recognised radio journalism and inserted them in the middle of the document I was working on.
I also have to be careful not to exhale noisily otherwise it inserts an unwanted “this”. (But it ignores a cough.)
And of course I can’t now listen to music while I’m working but I’ll have to get used to that!
The voice used for reading aloud does sound very mechanical though. I imagine that speech synthesis has moved on since the speech first program I saw on a Appple III in the early 80s, but it doesn’t seem all that much better.
As an aside, I remember when I tried that early speech system I tested it with an example invented by George Bernard Shaw who was a campaigner for rationalising English spelling and wanted to show just how illogical it is. How do we pronounce “ghoti”? Why, “fish” of four of course! gh pronounced as in “rough”. o pronounced as in “women”. ti pronounced as in “nation”.
The programmer of that speech software was way ahead of me! When I entered “ghoti” to my utter amazement it said “fish”.