Speech Recognition Software Still Has A Way To Go

Dragon NaturallySpeaking was popular, but also functioned poorly.

In the 1990s, there was a promise that speech recognition would become mainstream. We would be able to just talk to our computers, our voices would be perfectly recognized, and our essays would be completed before we could even finish talking.

According to PCWorld, Dragon launched the first consumer speech product, Dragon Dictate, for $9000 in the early 1990s. Seven years later Dragon NaturallySpeaking came out and allegedly recognized continuous speech, allowing you to put out 100 words per minute. You did have to train the program for 45 minutes though.

I remember experimenting with this in my classroom during my public school teacher days. It was about 85 percent accurate with my voice or the voice of some of my students. However, in order for voice recognition software to be successful, it needs to be about 98 percent accurate.

In the 2000s, both Windows and Mac OS X operating systems included voice commands in their operating systems, but they didn’t work well. Besides, most users didn’t even know the capabilities existed. For most of the 2000′s, speech recognition advancement seemed dead.

Apple changed the course of voice recognition with Siri, which first appeared on the iPhone 4s in 2011.

However, in 2011, Apple made major steps with Siri, the voice assistant on the iPhone 4s, which was released in October of that year. It was the first somewhat capable digital voice assistant to ever exist, at least in the consumer world. You could say, “Schedule an appointment on January 14 at 5 p.m. to go to the dentist!” and your calendar would have it set.

Many consumers have never previously seen or heard of something so accurate. However, there were problems when trying to operate Siri in somewhat noisy environments. People with thicker than usual accents also had problems. However, updates have made Siri even more accurate. To make things better, Siri has a personality. Just ask her quirky things like if she wants to make love to you and she’ll have a hilarious answer.

Google Now, which is Android’s answer to Siri, isn’t as accurate as Apple’s voice assistant,  but it’s close enough. Both Siri and Google Now are more than 90 percent accurate. It’s definitely a step in the right direction.

Unfortunately, it’s still safe to say that voice recognition hasn’t replaced actually typing in numbers or words for directions, taking notes, and other things. However, five years down the road, voice recognition software could be so good that we won’t even need keyboards anymore.

About Daryl

Daryl Deino has been a technology enthusiast since 1995 and has written for several newspapers and technology sites. Please reach him at [email protected]
This entry was posted in Tech News and tagged , , . Bookmark the permalink.

Leave a Reply

Your email address will not be published. Required fields are marked *


You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

Spam Protection by WP-SpamFree