Speech Recognition: Yesterday’s science fiction is today’s reality
This blog post was written by Daniel Hubbell, Senior Marketing Communications manager at Microsoft. Daniel’s career spans more than 12 years at Microsoft and his current role is focused on increasing the awareness of accessibility with consumers and regularly speaks at events and conferences on the topics of Accessible Technology.
Speech recognition is one of my favorite things to demo. There is something about watching words appear on a screen after speaking them that feels like science fiction to me. The technology has a wow factor that no other demo seems to produce.
Over the last 4-5 years I have demoed Windows Speech Recognition in some very unusual places. One of the most memorable of these was on a flight to Chicago where I impressed several rows of passengers. My demo started after a passenger across the aisle from me questioned my claims that the noise of the aircraft wouldn’t impact the results. I won the challenge, and had a dozen people applauding at 35,000 feet. The secret to my demo’s success was that ambient noise can mostly be overcome if you use the correct type of digital noise cancelling microphone, readily available in stores or online, for a reasonable price.
Recently, during a two week road trip to the International Technology and Persons with Disabilities Conference (CSUN), a colleague of mine and I stopped at a senior center to demonstrate speech recognition. My demo went without a hitch and, fortunately for me, cameras were rolling to catch the moment. I started the demo in front of a room of more than 100 by reading a paragraph from the newspaper. During the demo I was primarily concentrating on what I was reading and the commands I needed to give, so I hadn’t noticed the crowd’s reaction. I was very pleased to see in the following video clip evidence of how speech recognition can wow an audience.
Outside of work, my kids are no strangers to technology and given that, they are not easily impressed. They have been growing up in a world where carrying a computer in your pocket is commonplace. However, my most recent demo using speech recognition on my Windows Phone even impressed them. Windows phone 7.5 has the ability to send messages via SMS or Facebook chat by using your voice to dictate instead of typing. While driving one day I received a text message via Facebook Chat and because my car is equipped with integrated Bluetooth, my phone automatically read the message to me through my car audio system. After hearing the message and without taking my hands off of the steering wheel or my eyes off the road, I said the word “reply” and spoke my brief message back. After confirming the message translation, which had again been read back to me by the phone, my utterance of the word “send” sent the message on its way. My seven year old son broke the subsequent silence from the back seat by simply saying, “Man, that’s so cool dad.”
It’s wonderful to see speech recognition being used in such varied and interesting ways across Microsoft products. For more information about speech recognition in Windows, Windows Phone, and Kinect for Xbox 360, visit the following links:
