The built-in Mavericks OS X speech recognition is good piece of Voice recognition software to give you a taste of what speech recognition is like, but not as good as Dragon Dictate, mainly due to the complete lack of any correction capability. Here’s how it compares to Dragon Naturally Speaking, and how to enable the ‘Enhanced Dictation’ Mode which makes it a lot faster.
If you buy Dragon Dictate it comes with a built-in Microphone that will do the job, but if you buy the download version you’ll need to buy your own microphone.
Here are some of the options you have. Continue reading »
With Dragon NaturallySpeaking, the only option was to purchase it in Australia because there was a special Australian edition which was made using the Australian accent.
With Macspeech Dictate however, the US version comes with Australian accent, so you can buy the normal version of MacSpeech Dictate from the US for about $200 ($100 less than Australia) and it will still work here.
It comes with USA, UK and Australian versions built-in.
You can read my full review here.
Other posts on Speech recognition:
EARLIER REVIEWS on MACINTOSH HOW TO
When Macspeech Dictate was first released I gave it a negative review. I have left the reviews of earlier versions here as a reference – you can read them below….
The most up to date review is here.
28th April, 2008 Macspeech Version 1.0
Ever since I first used Dragon Naturally Speaking Version 4 in 2000, I have been waiting for it to be released on the Macintosh. Well it’s finally arrived, it’s called “MacSpeech Dictate”, and it’s powered by Dragon. Unfortunately there are still a few hiccups. My advice is if you have been using voice recognition already – Viavoice or iListen, then make the switch – it’s an improvement. But if you are not currently using speech recognition, wait a little longer. It’s not quite there yet, and the adjustment to speech recognition combined with some bugs makes for a steep learning curve.
Now in terms of dictation and recognition, MacSpeech 1.0 Dictate is fantastic. The accuracy is very impressive and it does not take much training. In fact I think I was reading stories for about five minutes and then I was ready to Dictate. It’s also a very stable application, unlike IBM ViaVoice which was frequently crashing.
It integrates very easily so that you can dictate into any Macintosh program. IBM ViaVoice was meant to do this but in reality whenever you were dictating into anything other than the ViaVoice speakpad it was tediously slow.
So what are the problems in 1.0?
Firstly there is a bug in it so that if you correct anything with your mouse and keyboard, it gets out of sync and loses it’s place, and you can’t keep dictating.
Secondly, it doesn’t train itself from your mistakes.
One of my favourite features of Dragon NaturallySpeaking was the correction window where the program automatically learnt from your mistakes. When it mis-recognised a word, all you had to say was “correct that” and type in the correct word, and Dragon NaturallySpeaking would automatically train itself so that whenever you spoke that word again it would get it right.
When Macspeech Dictate makes a mistake it just keeps misrecognizing the word, and you can’t re-train it. The manual promises that this correction feature will come in an update – let’s hope it’s soon.
Conclusion – Macspeech Dictate Version 1.0
If you have been using ViaVoice or iListen, then you will be blown away by the accuracy of Macspeech Dictate – it’s simply the best voice recognition software for the Mac.
On the other hand, if you have come across from the PC using Dragon NaturallySpeaking you will be taking a step backwards due to the lack of automatic correction, and the buginess of the software.
It’s a very nice piece of software, but with its current lack of features and bugs, it should be a beta version. In my opinion, it’s worth $100, $200 if you really need it, but at $300, it’s way overpriced given it’s problems.
NOTE: All these features that I mentioned needing improving have been fixed in version 1.3 and even more features have been added to 1.5 Macspeech now automatically learns from it’s mistakes so it trains itself as you use it. Also editing your text after you have dictated is much easier. And it’s much more stable.
Other posts on Speech recognition: