Microsoft did not want to mention the weakness speech recognition win10 ppbox

Clothing & Fashion

Microsoft did not want to mention the weakness: Win10 speech recognition Tencent digital news (Wen Xin) according to the PCWorld website, Windows has a function of Microsoft is not willing to mention. Although Windows allows users to use the stylus "paint", with the help of Windows Hello using a face login system (or Web security protection), even ordered Huna set reminders, a function but it clearly does not want the user to use, the system allows the user to edit the document or issue orders left and right by using the speech recognition speech input engine. Why Microsoft not vigorously promote Windows voice recognition function can be traced back to 10 years ago, when Microsoft product manager Shane · bouche (Shanen Boettcher) messed up in the voice input function demo Windows Vista when. After Windows voice input technology has been quite low profile, there are almost no users know that the Windows voice input function. If the Windows has a chance to play again the voice input field, now seems to be a time for progress of computer and artificial intelligence provides the basis for much better for voice input. Asked about the future of voice input technology in the Office, Microsoft is responsible for speech recognition research and the executive vice president of Shum, and the need to be, said · (Harry), this is a major problem. Voice input does not play a more important role is not understandable." Speech recognition is not perfect part of users still believe that speech input or "dounce Billy" (Doonesbury) series of comic Newton apple PDA, the user said "I am writing a test sentence", "Siam fighting atomic sentry to". Users have this idea can be forgiven: Windows voice recognition feature is still using Microsoft Speech Recognizer 8 technology, since Vista, the technology is basically no change. It’s called "grandfather" technology. PCWorld said, but the hardware has changed a lot: the ability to listen to and explain the voice of the processing capacity is far less than 10 years ago. Surface Book and other PC products in the quality of the integrated microphone array means that without the use of special microphone speech recognition can achieve high accuracy. However, the development of voice input technology has been suitable for the public to use it? When using voice input software to enter an article with a length of 1028 words, the accuracy rate of 95% means that the user must correct more than 15 errors. In the test, the Windows voice input accuracy of 93.6%, in theory, this value is not high, lower than the test of other dedicated voice input software. Windows has a strange habit of inserting "comma" into the document when a comma is entered. Voice。相关的主题文章: