This month Macspeech have released a major upgrade of Macspeech Dictate and they have also renamed it to include the Dragon branding – Dragon Dictate.
I’ve only been using it for a couple of hours, but I can say it’s definitely an improvement on MacSpeech Dictate. Dragon have put their name on the product, which I assume means they’ve put out something that they’re finally happy with.
I was happy with the last version, but this one is even better.
Dragon Dictate’s remarkable accuracy is powered by the latest version of the Dragon speech recognition engine. This is the same state-of-the-art technology used by Dragon NaturallySpeaking 11 for the PC.
All the previous upgrades to MacSpeech Dictate have just been an upgrade to the application, but this upgrade includes new speech files as well. That makes it a hefty download, over 1 GB, and also means that you have to totally retrain your voice from scratch. That’s not a problem though, because after no more than 5 or 6 minutes of training my version of Dragon Dictate is already more accurate than the last MacSpeech Dictate, which I had done the full training on.
Dragon Dictate, the latest version, also has voice controls, like ou can move the cursor around on your screen, and you can even say a phrase like ‘search google for Italian restaurants in Dubbo’ and it will open google and do the search. I’m not sure that I’ll use this feature a lot, but it certainly will be good for people who want to use DragonDictate to control their computer rather than just to dictate documents.
Note that Dragon Dictate only runs on Snow Leopard.
Here is a short video I put together of me trying out the latest version of Dragon Dictate…
Here is me dictating the start of a talk so you can see how it works in real life…
Here’s a video of what it looks like to do the basic training… (for some reason the audio doesn’t work for the first 30 seconds of the video.)
As good as speech recognition is, sometimes it drives you bonkers. Here it struggles to correctly recognise the word Australian – a shame since it is the Australian accent version!
Here’s a good little demo I found of an American guy doing a similar demo.
Other posts on Speech recognition: