Talking To Your Computer

From the February 2012 digital issue.

When Armageddon comes, I am convinced that it will not be the result of an epic battle between good and evil, or even between the haves and the have-nots. The battle that destroys the known universe will be, as it has been for several decades now, between Microsoft, Apple and Google.

Everywhere you turn, you are confronted by the armies of these three giants, scrambling for another point of market share and crushing or buying any third parties that get in their way. And now these three forces have found a new battleground in software that enables voice recognition and commands.

You know, the software that lets the crew of Star Trek simply speak in English to search databases and execute commands. There are essentially four players in this tech area:

• Microsoft, which is scrambling to improve on their lead in this market. Microsoft has bundled free speech recognition software with its operating systems since Windows XP, improving the system year after year (hint: you’ll find it in the “Accessories” folder). For the record, the Windows 7 iteration works amazingly well. In fact I’m dictating this paragraph via the Speech recognition program. If there’s a weakness to Microsoft’s implementation, it is that editing capabilities are very limited.
• Apple has built some limited speech recognition capabilities into its OS X operating system to assist in navigation. For dictation and editing, you will need third party “Dictate” software from Nuance. But this is just the tip of the iceberg for Apple, which has built some fairly strong voice command capabilities into the iPhone 4S, and has pushed a new, voice-driven application in its “Siri” personal assistant application.
• Google has no intention of being left behind in this new technology. The company has bundled voice capabilities for search into its Chrome browser, and offers robust voice command and search capabilities into the Android operating systems for cell phones.
• And then there is Nuance. This company is best known for its Dragon Naturally Speaking software, which cropped up in ads during the last holiday season. It is sharp, professional-level program whose only limitation is its learning curve, which is somewhat higher than that of the other products. In addition, just before last Christmas the company purchased a cell phone app maker called Vlingo, an up-and-coming player in the cell phone voice command arena with a decent app that works on Apple, Google and Blackberry devices. Whether Nuance becomes the fourth major player in this cat fight, or is simply absorbed by one of the other three, the battle is shaping quickly.

It would be nice to point clearly to a winner among these contestants, but that isn’t going to happen this year.

In the cell phone arena, Vlingo and Siri are competing with in-built voice command systems, and none of them do it well. Siri is that the moment just the personal assistant app it claims to be, with virtually no editing or message handling capabilities. Vlingo has excellent features to read and write messages, but does a horrible job with simple phone calls.

On the desktop, Dragon Speaking Naturally takes top honors as a general speech recognition and command system, but comes with the highest price tag. Microsoft Speech Recognition is free, but stumbles in a number of areas. It correctly recognizes the word “colloquialism,” but doesn’t handle “ain’t” and “y’all.” Google’s system requires the user to use the company’s Chrome browser, a program whose main function is to track everything you do online and sell that information to advertisers.