Google is using Deep Learning algorithm for voice recognition and translation. In recent years, we can see that voice recognition rate and translation level have improved greatly. In fact, in the past, when translating Hangul into English, it became widely known that translating Korean into Japanese and then translating Japanese into English is a little better than translating Korean into English directly. I did not get that satisfiable level of results so well that I would talked to myself “I would have translated from the beginning …” during proofreading.
Nowadays, Google translation gives me surprise with the cool translation output. When you use translating Korean into English with Google Translator, you have to go through the correction process once. These corrections will identify situations where recognition varies depending on particular homonyms or contexts issues. Most cases, most problems can be easily detected and corrected.
When I was giving a training on the teaching and learning center of Changwon University. Unfortunately, the notebook set up at the computer room could not show the demonstration because it did not recognize the microphone properly. So I came home and tried voice recognition using my laptop in the same situation. If you look at the content of the video, you can see that you are now able to trust and use speech recognition and Google translation. This article was written using Google’s speech recognition and translation. Even though most part of the video is in Korean., you can understand it suport various language well in Goolgle. I tried to speak in 4 different languages. 🙂 Happy Googling ~