PowerPC to Core 2 Duo Migration Has Big Payoff for Dictation Software

For the most part, I remain very satisfied with the performance of most tasks I perform on my middle-aged 1.33 GHz 17-inch PowerBook G4, which I bought back in 2006. But not when it comes to dictation. Interestingly, the most dramatic performance boost I’ve realized transitioning from PowerPC to Intel Core 2 Duo is with dictation software.

I’ve been partially dependent on speech recognition since the late 1990s, when chronic polyneuritis flare-ups began to make typing more than a paragraph or two painful to the point of inhibition. Not that the first speech recognition software products for the Mac didn’t inflict their own sort of pain. Articulate Systems’ Power Secretary and its successor Voice Power Pro were based on discrete speech recognition engines requiring you to pause…between…each…word, an unnatural mode of articulation that tended to put a damper on the creative flow.

Continuous speech dictation came along for the Mac with IBM’s ViaVoice Millennium Edition in 1999, and since then speech dictation software has had many incarnations with many speed, stability and accuracy improvements. But even on a fairly powerful G4, like my PowerBook with its 1.5GB of RAM, it still always felt like the computer was working hard to keep up. And that’s because it was. Speech recognition software used for dictation is more complex and demanding of hardware power than most categories of consumer software, excluding perhaps sophisticated gaming and high-end video editing, and the G4 processor was never fully up to the job.

Core 2 Duo in Its Element

However, the potent processing muscle of the Core 2 Duo, which is frankly overkill for most of my computing needs, is right in its element dealing with the processing demands of dictation running MacSpeech’s latest Dictate 1.3 software. Dictate is an Intel-only application based on the superb Dragon NaturallySpeaking speech engine and, even on my modestly powered 2.0GHz unibody MacBook and its standard 2GB of RAM, it feels like it’s keeping pace without breaking a proverbial sweat.

Effortless, Enjoyable, and Even Fun

Dictation has always been a useful and helpful tool, especially for those of us with physical disabilities, but the combination of Core 2 Duo power and MacSpeech Dictate (which is the only dictation application still available for the Mac, IBM having bailed out of the category in 2003), makes it effortless, enjoyable, and even fun — worth considering even if you can type for hours without a twinge of pain. And while dictation is Dictate’s main raison d’ĂȘtre, it also supports computer navigation and control by voice, which can be vitally helpful for folks with severe typing disabilities, and potentially handy for any user.

Dictate requires “training” to create a user recognition profile, but that process has been streamlined and shortened to the point where I’m finding after an initial 5-10 minute training session, I’m getting better accuracy than I ever achieved with the old iListen program. I intend to do more training using the program’s built-in voice training tools, and to upgrade the RAM on my computer as well, both of which should polish and improve Dictate’s performance even more. But in the meantime, the program is giving me amazingly good accuracy and speed virtually out of the box and on standard spec hardware.

Leap of Faith

Purchasing a dictation application requires more of a leap of faith, so to speak, than with most software categories, because free demos aren’t really practical and a high-quality microphone is a must for achieving acceptable performance. Dictate isn’t cheap, selling for $199 (that includes a high-quality Plantronics headset microphone), but I don’t anticipate that many purchasers will experience buyer’s remorse, and if you use your Mac for serious production work involving a lot of text entry, the program should pay for itself fairly quickly in terms of increased productivity. Once they get used to it, most people can enter text much quicker via voice than with a keyboard, and it’s less fatiguing.

If you’ve perhaps experimented with voice recognition software in the past and been underwhelmed, you’re in for a pleasant surprise.

You're subscribed! If you like, you can update your settings


Comments have been disabled for this post