Technology on Command
Students Explore Speech Recognition Software
There’s no way around it: kids need to learn how to type, and not with one finger. Texting doesn’t count. However, it’s not hard to imagine their world as adults when their voices will do much of the work and the common desktop computer with a keyboard will be left behind as they talk into their watch-sized computers.
Speech recognition software has already made huge advances since it was introduced more than two decades ago. Recently, students at Empow had a chance to test out the latest technology when representatives of the Burlington, Massachusetts-based company Nuance visited Empow Studios. They demonstrated their Dragon Systems software, which was originally released in 1994. The technology has come a long way and allows users to dictate their speech and capture user’s intent through Natural Language Understanding (NLU). Newer software offers close to real-time dictation and works in numerous desktop applications. It can also be found in mobile devices, automobiles, and in smart home devices like refrigerators and thermostats. Speech recognition is being used to make the lives of families easier, and also more accessible for people with disabilities.
During the presentation, Dr.Naveen Parihar, Ph.D, a member of Nuance’s core speech recognition engine team, explained how the challenge of recognizing a human voice is met using a ‘divide and conquer’ approach that is fundamental to many computer science algorithms. “Given an audio signal, the speech recognition process is broken down, based on expert knowledge, into components like signal processing, language modeling, acoustic modeling and search,” says Dr.Parihar. “The results of these components are combined to arrive at a sequence of words.”
The students were impressed with the ability of the speech engine to understand and track their words. “I noticed during the presentation that the kids were constantly asking questions,” says Empow administrator Taylor Tutwiler. “They were fascinated by the presenters because they had helped create the technology they see everyday on their parents’ phones and tablets.”
“It was nice to have a technologist with deep research and development experience in the area of automatic speech recognition talk about his everyday life at Nuance because the kids here start relating what they’ve learned about coding and how they can apply it to cool innovative technologies,” says Empow outreach director Dave Gutierrez, who strives to boost learning at Empow with varied technical demonstrations. Of course, there’s always elements of fun. “One of the kids asked, “Dragon, play a rock song,” and Katy Perry started playing,” laughs Gutierrez. “Everyone wanted to try it!”
“The students have curious minds wanting to learn and we are eager to share our passion for speech recognition,” adds Dr.Parihar, who says he also benefitted from the event. “We enjoyed interacting with the students. It expanded my understanding about what is important to them, and how much the world has changed since I was in school. As always, learning is bi-directional.”