Speech recognition was indeed developed long back by the tech giants, but the innovation was not popular as the software did not perform well for the different accents of different people in different countries around the world. But now, a research team has finally come up with a 'Speech Recognition' technology which is 3 times faster than human typing in smartphones
According to Stanford News, a research team with computer scientists from Stanford, Baidu Inc. and the University of Washington conducted an experiment with Baidu's Deep Speech 2 cloud-based speech recognition software and 32 fast texting people aged between 19 and 32 with Apple's built in keyboard in their hands.
James Landay, a professor of computer science at Stanford and co-author of the new study said "They grew up texting, so we're putting speech recognition up against people who are really good at this task,"
The subject used to compete the speech recognition against the fast typing were taken out from a standard library with standard phrases like "physics and chemistry are hard," "have a good weekend" and "go out for some pizza and beer". The time was clearly recorded for both the parties and the experiment conducted to test the Baidu's Deep Speech 2 cloud-based speech recognition software against the human typing and came up with best results.
The experiment was conducted in English using QWERTY keyboard and Mandarin using iOS' Pinyin keyboard.
English language speech recognition gave appreciating results showing that speech recognition is three times faster than typing with error rate 20.4 percent lower. Whereas Mandarin Chinese is 2.8 times faster than typing with error rate 63.4 percent lower.
According to co-author Sherry Raun, a computer science PhD student at Stanford said, "We knew speech recognition is pretty good, so we expected it to be faster, but we were actually quite surprised to find that it was almost three times faster than typing on a keyboard,"
Landay said "We should put speech in more applications than just typing an email or text message," He also added, "You could imagine an interface where you use speech to start and then it switches to a graphical interface that you can touch and control with your finger."
Stanford experiment shows speech recognition writes texts more quickly than thumbs